November 30, 2011
HP will migrate best of HP-UX to Linux x86
No problem, HP says, if you're an HP-UX customer who's just purchased an Integrity server to replace anything else -- whether that's from an HP 3000 or just an older HP Unix system. The HP Odyssey project is not a signal of any new future for your Business Critical Systems server choice.
The popular belief among customers -- who still ask HP about it -- as well as the market and resellers is that "Oracle is turning the screws on HP-UX, which is dead anyway, because Itanium processors are now too far behind x86s to ever catch up." Okay, that last part's not completely fair to the Itanium systems. The big-scale customers love their hardware and OS. Well, they did until recently, but it now remains to be seen how good the solution looks in the face of HP's traipse into the Odyssey.
In the LinkedIn HP-UX users group, Steve Shaw, Chief Technologist at HP Canada, says
HP-UX and Integrity support and development are continuing, so Project Odyssey is not "designed to get your HP-UX apps and systems to HPx86." Rather, it is designed to provide an option for apps to either stay on HP-UX, or migrate to mission-critical x86 (Linux and Windows) -- whatever is best for the customer/app/situation, all providing the mission-critical experience that's been delivered by HP-UX over the years.
There's an echo in the mind of any HP 3000 customer who's made this odyssey before, however. After a formal announcement of Itanium tech work for HP 3000 users, HP pulled back. Dave Snow delivered the news that the 3000 wouldn't lead the way in such servers. In 2000 he said it wasn't needed.
The 3000 division isn’t going to lead in terms of moving to [Itanium's] IA-64. We don’t need to. We’ve got good-performing chips that provide us with the 30 percent per year performance increase — maybe even exceeding that — for several years to come.
That was a promise that a current HP technology will get continuing development and support. You could swap out "Shaw" for "Snow" and feel the same vibe. True, there's no risk of an overnight exit from Integrity and HP-UX, not like the deal done on 3000 sites one year after Snow's comments. We just heard a story from STR Software's Ben Bruno about one spectacular dump of 3000s within days of the exit announcement. It resulted in a migration to Sun -- and some customers are figuring Odyssey will trigger the same kind of exit to another vendor.
It's not HP's support that's of immediate concern for any HP-UX customer. It's that Oracle vow to drop Itanium support. HP's been admitting that Oracle has pulled down HP-UX since this spring. Some customers see the start of the last mile of the Itanium road, and with it, HP-UX. A few weighed in to answer a question up on the HP-UX Users group of LinkedIn.
"The future for HP-UX looks bleak," said Rainer Grebin, listed on LinkedIn as the Technical Services Group Leader for the UK's River Island clothing and retail stores. "Unfortunately project Odyssey will ultimately drive most companies to IBM's AIX. [HP-UX] features on Linux are desirable, but Odyssey won't get many customers to migrate to Linux. I think it is very interesting that only Linux and Windows are supported. What about VMWare? Most customers choose software based on features and don't care for the OS or hardware platform. Most apps vendors do not certify Linux if they already have an alternative such as AIX."
It could be worse. Bruno's story details his Nov. 14, 2001 experience, when he was a reseller working on a deal with an Ecometry e-commerce site. "I was working with a brand-new Ecometry customer who had just paid for their twin HP 3000s just days before," he said. "I was promoting the benefits of our automated fax products that functioned the same on MPE (with FAX/3000) and non-MPE (with FaxCommander). Sadly, no one from Ecometry or HP even gave them a clue that the systems they'd just spent millions of dollars on were not a good long-time investment."
Ultimately, that e-commerce company decided to purchase a package on Sun's Solaris Unix, Bruno added, "and I don't believe HP or Ecometry refunded their money." It would be unbelievable if they did. HP's not exactly predicting an unbroken future for Itanium, once you get away from the Chief Technologist level.
"The x86 systems will rule over the Mission Critical environment and the reign of RISC will come to an end," said Mohannad Daradkeh, Technology Consultant at HP. "Bringing the Unix-like features into the x86 is a move in the right direction; hopefully the market will be mature enough to accept this."
HP's hope is that since HP is going to give to Linux x86 what makes HP-UX special, there's no reason to leave a platform that's sinking, according to CEO Meg Whitman. Engineers have their doubts, though.
Philipp Prokopets, Support Engineer at HP Enterprise Services, said on LinkedIn that "I'm really not sure that Odyssey will stimulate HP-UX and Itanium, unfortunately." News of an odyssey that brings the best of a proprietary OS to an industry standard like x86 can't be good in the long run for an Itanium investor. The best hope of HP-UX survival is going to be from a vendor like Stromasys -- now finishing a first release of a PA-RISC 3000 emulator, one that could serve the HP-UX customer nicely.
HP-UX to Linux is an easy move, according to Marxmeier Software's Michael Marxmeier. The man whose company has provided migration sites with a database that works like IMAGE doesn't see much concern about shifting away from HP's Unix, even with Itanium's future in play.
"Itanium certainly has its users," he said, "and it’s hard to tell if it will make it or not. However, this shouldn’t be a concern to the customer. They’re not pulling the plug on anybody. But if they’d like to move to something else, the proven technology of Linux is readily available. About half of our customers are using Linux these days." The Marxmeier database Eloquence "even supports some exotic implementations, like Linux on Itanium."
November 29, 2011
3000 team awaits one last strike, next year
By Ron Seybold
Scary and sad things can happen deep in the night. I learned that in Switzerland and again in Texas, both sets of news that arrived deep in the fall of seasons 10 years apart. But for each bad report, there’s the prospect of better news for a season to come.
The first scary news arrived on a pay phone in a rail station. It was in a November night beyond 9 Central European Time, back in the days when Daylight Savings ended by October. I’d already thrilled to getting the news that the Yankees lost a World Series in a Game 7. That’s the kind of news that can cut both ways, but it would take me another decade to learn that lesson.
The news on that Swiss night was that HP wasn’t going to build any more HP 3000s in 24 months, that they believed everybody ought to get off the platform. That Unix was the best solution, or Windows, anything but what you knew, built your business around, slathered all over your future, your training and career. It was damp and cold on that railway platform hearing that news. My boy Nick and I were on our way uptown to Lausanne and dinner. The report from my partner Abby Lentz sapped my appetite. I did my best to explain to my son things were changing for my business, but it would be okay. Sometimes there are things you just have to say and wait for them to become true.
At no time that night did I ever believe there would be another decade of work on the HP 3000 for my family. Ten more years seemed impossible on that night, in that month, anytime over the next few years. It seemed as impossible as being unable to get one last called strike in a World Series, twice, 10 years later. That happened far into a dark and cold night, too. Past midnight in a cold, damp ballpark in St. Louis.
But for every leading home run that my Texas Rangers could hit in an epic Game 6, their opponents the home team Cardinals could avoid that last strike my guys needed to win their first Series, ever. Not even the extra innings “Roy Hobbs homer” from straight-edge hero Josh Hamilton, swung out over a sports hernia that required surgery, could power the Rangers into the champagne and champions’ ball caps. So in 11 innings, they — or sometimes we 40-year-fans of baseball, we say “we” — lost that chance to win.
But just like the HP 3000 community, those Rangers have not lost all chance ever to stay in the game.In sports, after we lose we like to say, “There’s always next year.” Here in Texas we believe it about the Rangers, turned into champion-caliber players by the legend Nolan Ryan, now an owner. And in my house we believe it about the HP 3000, too. There will be a very interesting next year, a 2012 with an emulator that puts 3000 hardware onto speedy PCs makes its debut. It’s the kind of news that’s sparking sort of a “hot stove league” among people still using HP 3000s. Hot Stove is the time before the season starts, the in-between after a Series and before Opening Day. A million different questions and scenarios and combinations get kicked around, and it’s called Hot Stove because it’s cold almost everyplace people care about baseball.
Except in Texas in mid-November while I wrote, as the clock and calendar ticked over into November 14, it’s 70 degrees and the windows are wide open, even at 2 AM. Things are not what they used to be in our world. Summer brought 5 inches of rain here. Up north the blizzards were followed by floods. Sometimes doing this job means working until it’s already mid-morning in Europe. Like all of the 3000 veterans and experts who told me moving, fabulous November 14 stories, I’m just putting one verb in front of one noun, like they’ve put one consulting gig in front of one temporary contract. We’re all trying to stay active while we’re in the batter’s box, waiting on whatever pitch we will see next.
People still care about the HP 3000, even if they’ve left it behind for something mandated by management or dictated by datacenter needs. Phrases like “the machine I hold so dear” and “it’s still right at my side” are what flow from tales of how you’ve grown over the years. They’ve been hard years, in some places for some people, and they’ve also gotten people involved with new passions and lessons. Experts of 30 years of MPE say they’ve learned new tech like Ruby on Rails or open source security, and found it fun. My partner Abby, still dreaming up NewsWire concepts as publisher, gave birth to a yoga practice that’s produced two DVDs. Me, I learned to write and teach fiction, the drama of journalism grown richer, written to move the soul without excuses and no rebuttals. I always wanted to do that, and HP spurred me dig in and learn. The world I knew was changing, like yours. I had to add another dimension to my writing game.
It’s a lot like what those Rangers of mine face during these darker and cold off-season months. Josh’s hernia will heal, the young team will rest up after 178 games and come back with a new dimension: being just one last strike away from winning the last game of the season. And when spring arrives and weather warms to the desert we’ve come to expect in Texas, there will be a fresh chance to win. Like the new season for the 3000, building upon its community and its deep IT experience, and now with a new dimension of virtualized hardware and source code licensed to top support shops and developers.
When you’re only one strike away, you’re close, as close as I am to finishing that first novel of mine. Whether it’s playing with words, or balls and strikes, or the magic of computers built out of just bits on a disk, the next season, story or release brings more hope. After 16 years of playing on this newsletter’s field of dreams our sponsors and readers helped us build, Abby and I can be glad this stove remains hot, while we get another swing at our joyful pastimes. We’ll see you here in print again in February, when it will be time to start to play ball, buy an ebook of mine, and boot up a fresher future.
November 28, 2011
Making it Easier, One Decade Later
The tools of my trade have come a long way in a decade of work, just as yours have in computing. Actually, there's an intersection there between us, since the tools of computing have changed the way I can tell stories. Journalism is sometimes called literature in a hurry, and there's nothing like the rush of big news to put a scribbler in a rush.
That's what was happening in November, 2001, when I got my rock-your-world report from my partner Abby about our newsletter and the HP 3000 and the future of both. We knew HP's future in the 3000 that night, when it was likely to wind down and how serious they were about ending it. We also knew there was another side of the story to tell about the community that the vendor was leaving. HP was already at work telling their tales about migration and a declining ecosystem. We had to get busy to catch up.
It all seems so antique now. The long ride on the train with an open notebook on the Eurostar table, writing out question after question. Getting a hotel with a good phone system booked in London, for HP had promised a con-call to brief me less than a week before they'd tell customers. Shopping for a portable cassette tape recorder, plus an acoustic mic pickup, to make recorded notes for such a crucial story.
Then afterward, at 8 PM Euro time, when it was just lunch in California, I trek off to the EasyNet Internet Cafe. A spot that any tourist or pilgrim could use the Internet to write stories I would email to Abby, via AOL, so she could set them in print for our newsletter. We'd held the presses, yup, and it all depended on how fast I could get quotes out of that tape recorder and its cheap earphone, writing up the news I'd heard, and then another piece on what I thought it meant.
I got lucky that night. There was a problem with the Cafe's billing, so everybody's Internet time was free. It was one less thing to think about, so I could let the muse lead me to call the non-migrators "homesteaders." When you create something that's swell and durable, you must thank the gods for good fortune that rewards the practice your craft.You've probably felt that kind of luck in creating something during your careers -- that insight that turned out to be right and make the whole program run without any error on the compile, or the magic of making connections between two things the makers never intended to work together. You try to remember how it felt, because you want to believe in the luck for the next time you're batting at the keys in cafe where it's about to snow outside and you're not dressed for it.
Skip ahead 10 years, and you don't worry about which COBOL to use or whether that's the right SCSI cable and terminator. Or if you've got time on the Internet or the hours to skim through tape for one quote. As I tapped out a blog story about the great Steve Jobs, how he admired Hewlett and Packard, and stop that HP PC snickering, he said very near his own demise. They built a company with a legacy, and he wants that for Apple.
I'm right in the middle of this, when HP calls a pop press conference with its new CEO. Decides that PCs are still its game. I listen right along with every other business writer, get a story posted online along with the Merc in San Jose and the Journal. Just luck along with tools that make stories leap through time, another kind of Easy Net. Thanks for the work you've done over the last 10 years in your industry. It's making it easy.
November 25, 2011
HP relents, aims Unix sites at x86 futures
It wasn't exactly an announcement while you were sleeping, but HP slipped in the news of its Unix migration during the very slow Thanksgiving business week. For some customers it might be a turkey, and for others who want to get onto Unix for the short term, or stay there and see an HP future, they may be thankful.
HP will start development of a next-gen platform for the future of HP-UX users, one that runs on Intel's Xeon x86 chips, the company announced Nov. 21. HP's Unix will continue to run on the almost-proprietary Itanium chips inside Integrity blades and servers -- and as recently as September HP said it wouldn't port HP-UX. It's Unix is still not making the x86 jump, but the things that make HP's Unix special will make a leap to Linux and Windows. There might be a lot of Unix application rewrites going on, although the GM of the HP Unix unit, Martin Fink, will insist otherwise. For awhile.
HP's new CEO, the need to halt the slide in the Business Critical Systems sales, and flat HP sales overall must have done the trick. "The BCS business is a declining business," said CEO Meg Whitman last Monday in Q4's briefing. "It is a slow decline, but I don't think you're going to see an accelerating growth rate in that business. And so we just have to manage that as best we can and invest in R&D so we get to a new platform as fast as we possibly can that allows us to service the clients that need this kind of power."
The project that HP is calling Odyssey will "redefine the future of mission-critical computing with a development roadmap that will unify Unix and x86 server architectures, to bring industry-leading availability, increased performance and uncompromising client choice to a single platform." What this amounts to is the migration path away from HP-UX to Linux, something many of HP's Unix customers have seen in their futures for some time. An intensive article with the tech details of how HP's going to leverage its latest Itanium sockets into x86 is online at The Register. "We are systematically evaluating the arsenal of intellectual property for HP-UX and mapping that to x86 platforms," the article quotes the BCS chief technologist.
"Some of the technologies that make HP-UX and its Integrity and Superdome platforms rock-solid will end up being donated to the Linux kernel," the article adds.
What's not stated in that HP press release is the future of Itanium in the HP server line. A "single platform" signals that the Intel Xeon chip family is going to win out in HP's near future. This probably means the end of the Itanium developments from Intel, after its next two processor rolls become a reality. Perhaps sooner, if the demand for Itanium servers keeps sliding.
For the next two years Itanium is likely to remain as key to HP's Unix as it has during its declining period, if the history of bringing an HP OS to a platform holds up. For more than three years during the previous decade, HP worked on getting HP-UX to run as fast on Itanium as it did on the company's prior proprietary chips, PA-RISC. In that case there was more development going on inside the chips than in the OS. This time around it's HP's primary job to make its Unix distinctions a choice for the more common and popular x86-Linux line.
Bad news for the users of the other Itanium-locked environments, however: NonStop and OpenVMS transitions to x86 are not part mentioned in Project Odyssey. The language to be teased apart in the announcement says there will be "innovations to HP Integrity servers, HP NonStop systems and the HP-UX and OpenVMS operating systems." But only Unix customers will be given a path for their enterprise advantages onto the industry standard (read: x86) among HP's environments. HP is likely to tell NonStop and VMS users they won't need such industry standard pathways to keep their servers growing and on par with the x86 line. Itanium is going to go far enough.
HP 3000 customers will recall a similar stance from HP back in the earliest days when the vendor hawked Itanium futures. We all know how turning MPE away from Itanium turned out, although the difference is that the VMS and NonStop lines have more users than MPE/iX did in 2001. Both VMS and NonStop have a lot fewer installations than HP's Unix, however.
"Customers need the availability and resilience of Unix-based platforms along with the familiarity and cost-efficiency of industry-standard platforms," HP said in a statement. HP specifically cites Oracle as a heavy anchor to Itanium sales. This Odessey move has a lot to do with Oracle's pullout from Itanium development. Although HP is suing Oracle to back off of that plan, getting Unix users onto a path to x86 will let HP continue to sell HP-UX servers as Oracle servers -- until the Odyssey is complete.
Martin Fink, the GM of the beleaguered BCS unit -- its sales dropped 23 percent in Q4 -- said that customer demands are driving this odyssey.
Clients have been asking us to expand the mission-critical experience that is delivered today with HP-UX on Integrity to an x86-based infrastructure. HP plans to transform the server landscape for mission-critical computing by using the flexibility of HP BladeSystem and bringing key HP technology innovations from Integrity and HP-UX to the x86 ecosystem. Unlike the competition, HP offers an open, integrated, single platform approach.
Note that the word "migrated" is omitted from Fink's "innovations" sentence above. HP's got to do that migrating. HP-UX apps will be making their own transition, this time to a new OS (Linux) instead of a new HP chipset (PA-RISC to Itanium). There doesn't seem to be a HP-UX 11 v4 in the fortune teller's cards.
Project Odyssey provides assurance to current and new clients that HP-UX / Integrity environments will continue to be supported and enhanced through the decade and beyond. Clients investing in a mission-critical Converged Infrastructure today with Integrity and HP-UX, if desired, can evolve to a mission-critical Linux/Windows environment in the future.
The key to the HP strategy is going be moving those mission-critical and high-performance elements of HP's Unix to x86 hardware. That's going to translate, if HP can pull it off, to giving a Hewlett-Packard instance of Windows or Linux a leg up in a competitive market against IBM and Oracle -- even while HP says Oracle is running a distant third in the Unix server race of this year. HP's promises include:
• Increase scalability with 32-socket “DragonHawk” symmetrical multiprocessing x86 systems that will scale to hundreds of cores and support large, complex workloads. The systems will enable clients to deploy the smallest to the largest workloads in a dynamic, highly scalable pool of IT resources.
• Increase reliability and flexibility with two-, four- and eight-socket “HydraLynx” scalable x86 server blades with mission-critical virtualization and availability, all packaged in the robust c-Class enclosures of HP BladeSystem.
• Increase availability of critical Linux applications with the HP Serviceguard solution, which automatically moves application workloads between servers in the event of a failure or an on-demand request.
• Boost flexibility and availability of x86 systems with HP nPartitions technology (nPars), which provides precise partitioning of system resources across multiple or variable workloads. HP nPars is electrically isolated to eliminate failure points, which allows clients to “scale out” within a single, robust system.
• Enhance business continuity with HP Analysis Engine for x86 embedded into the system firmware. HP Analysis Engine goes beyond error logging to ensure efficient diagnoses and automatic repair of complex system errors while restoring system stability in seconds.
• Boost reliability and resiliency of x86 systems with fault-tolerant HP Crossbar Fabric that intelligently routes data within the system for redundancy and high availability.
There's not a word about HP-UX in the Odessy summary above. It's a lot of hardware that will not come close to serving HP's Unix customer.
So in a nutshell, that's symmetrical processing with a new 32-socket design of x86 systems, the DragonHawk line that correlates to the SuperDome systems of today. Then there's HydraLynx, which maps to the C3000 Itanium blade systems in the current line -- plus migrating Serviceguard, nPartions technology, the Analysis Engine and engineering the Crossbar Fabric into x86 systems. These last four items appeal to the large-scale enterprise customer who's been sticking with HP-UX.
HP didn't announce a timetable for delivering its new generation of Unix-migration x86 hardware systems. There's some talk of two years, and HP didn't cancel HP-UX futures or the Itanium hardware outright. But now the company gets to make an announcement, in order to stop the questions from BCS customers, about a transition away from its Unix. Mainly, that when a customer wants to make a Linux migration, HP will have consolidated hardware to do so, iron that provides the best of HP-UX features on Linux and Windows. Whether Oracle is going to keep its place running on HP-UX servers in the future -- no matter what Oracle decides to do, or be sued into doing -- is now going to about Itanium, not keeping the large-scale enterprise customer on HP iron.
Building up HP-UX hardware is pretty much over in the futures department. HP talks about retaining the best of HP-UX -- but not NonStop and VMS -- and building up Windows and Linux. There seems little way to view this but the admission that Itanium's got a limited development future at HP. If ever there was a message that says x86 isn't on the HP-UX roadmap, this path seems clearer than ever.
Apple ready to overtake HP in PCs
It's Black Friday here in the US, a day when retail outlets prod consumers out of bed with "doorbuster" prices that launch with a handful of rock-bottom-priced electronics. Silly things like $50 tablets jostle with $79 HDTVs and $329 HP 17-inch Pavillion laptops at Best Buy.
But Apple's going to be offering its gear at retail stores pretty much at list prices. The vendor insists its retailers don't do much of the come-ons for Black Friday, where 10 or so units at each store get sold starting at midnight. (Tickets for the BestBuy midnight entry were passed out last night). The retailers like it, but it can't do much for the vendor's sales at such small quantities. A report just released by analyst Canalys shows that Apple is poised to take over HP's position in PC sales soon, if not in the current quarter. That's the No. 1 position. HP's got no tablet to sell, while the iPad 3 is expected next spring.
Canalys reports the overall PC market is growing by 18 percent this year. The growth has come at HP's expense, considering how fast Apple is overtaking the vendor. HP doesn't need to be No. 1 any more than Apple ever did. But it needs profitability out of its PC business, something Apple hasn't struggled to attain in the last decade.
November 23, 2011
As holiday looms, research migration today
The US Thanksgiving weekend -- the only guaranteed four-day holiday all year -- will be upon IT departments in just a few more hours. Watch the "out of office" auto-messages pop up on your email replies. But if it's before noon on the West Coast, or just short of the traffic jams on one of the worst commuting days of the year, you can learn more about data migration from an MB Foster webinar today. (Thanksgiving is an October holiday in Canada, where MB Foster is headquartered.)
The vendor reminds us that it's been in the migration business for decades by now, since migrating data has been a principal enterprise there. "Data generated and consumed by applications is driven by multiple business requirements, which in turn are supported by business processes. The approaches taken to the migration depends on those requirements. Data migrations and associated tasks are usually performed programmatically to move and map old system data to the new system," MB Foster says in an invitation to a 11AM PST / 2PM EST webinar, Data Migration, Conversions, Integration.
It's that data that is the most crucial element on 3000s still running companies. At the end of the process is decommissioning, a service the vendor is ready to help migrators prepare for.
In this webinar we will outline database migrations, data conversions and data integration approaches. We will also be reviewing data challenges that could possibly impede a successful migration and decommissioning.
There's still time to register online for the 45-minute briefing.
We'll be taking tomorrow off, with a little bit of coverage on Friday, to celebrate the holiday here and give thanks for your constant interest and our sponsors' support. Have a safe and sumptuous weekend
November 22, 2011
HP vows to get its R&D back on track
HP made the price very clear for its pullout of tablets and WebOS: 91 percent of its Q4 earnings evaporated, according to the quarterly analyst briefing delivered last night. But new CEO Meg Whitman says the company will be getting back on track to better profitability and growth in the year to come. Mergers & Acquisitions are off the menu for the next year -- unless something extraordinary comes by, under $1 billion, in the software sector.
Yes, even a CEO who's determined to turn HP back to its invention roots can have moments of distraction. For the most part, Whitman was clear on what the company needs to revive: Research & Development.
"If we're going to get out of big M&A, we're going to have to get our investment up in R&D," she said. HP posted sales numbers to beat analyst estimates, but the profits slid from $1.9 billion to $239 million during the period. HP booked about $1 billion more in revenues versus last year's final quarter, $32.1 billion (and $127 billion for the fiscal year). HP's Generally Accepted Accounting Practices (GAAP) results included $755 million in a "wind-down of the WebOS device business for supplier-related obligations and contra revenue associated with sales incentive programs." The consumer side troubles run deeper than a dumped product, though. Printer sales, the firewall of HP's business, have also started to decline.
Then there was also a charge against earnings listed as "Impairment of goodwill and purchased intangible assets," $855 million. That turns out to be costs "associated with the acquisition of Palm Inc. on July 1, 2010 recorded as a result of the decision... to wind down the webOS device business." Whitman said HP's own actions impaired it during a quarter where it also toyed with spinning out PCs and axed its CEO. "One third of our challenges in 2011 were HP-specific," she said. "I had customers tell me they thought we were getting out of the hardware business entirely."
Another third of the challenges were "the distraction factor," she said. "There was a lot of drama in 2011." There was also a lot of silence on the cash register for HP's Business Critical Systems. BCS sales were down 23 percent for the group selling Integrity servers and the HP-UX, VMS and NonStop environments. These were the systems weighing down the overall results of the Enterprise Servers, Storage and Networking Unit (click above for ESSN details.)Those are not numbers that will spark Hewlett-Packard to launch a move of its HP-UX to Intel's other chipsets -- although that's just the kind of R&D that would fit Whitman's vision for the company.
"We cut out a lot of muscle in R&D," she said of a period she described as fiscal 2011, but analysts say reaches much farther back than that.
She said she's seen acquistions work to grow other companies, but the M&A mantra is going to be fading at HP. "We cannot rely on acquistions alone at HP. We have tremendous R&D capacity here."
R&D projects started now will not pay off in product before 2014, she admitted. It's just the nature of creating something internally instead of spending billions. Although she praised the just-closed Autonomy deal, an $11 billion venture. For the fiscal year, HP increased its long-term debt by about 50 percent, to $22 billion. HP is going to be repurchasing shares in the coming months, another sign of reorganization.
"We're building HP to last, to be great over the next decade," she said. "We just can't continue to run this company for the short term. We need to get back to the business fundamentals in fiscal 2012, including making prudent investments in the business and driving more consistent execution."
November 21, 2011
Coming soon: HP's year-end business report
HP wraps up its fiscal year in October, so its news of 2011 arrives today, three weeks after a tough year has closed. The fourth quarter results arrive in a webcast starting at 5PM EST. It's a period with a potential PC business pullout, the flameout of a tablet and a mobile OS, an ongoing war with its biggest database partner, and the ouster of another CEO.
Even with all those anchors dragging down the company, analysts expect HP to report flat sales -- equal to 2010's Q4 -- while profits are forecast at $1.13 a share. HP missed analyst estimates last time out, a quarter tougher than any other the CFO Cathie Lesjak ever reported. After shares fell into the low $20s, they're up 40 percent today trading around $28.
HP's business health means something to the 3000 site which has migrated to the vendor's other platforms, or plans to do so. But its Unix solutions are really becoming a choice for a very large customer. "HP has become a provider for the largest customers they have," said Michael Marxmeier, whose Eloquence database has exploited HP-UX at least as much as any supplier. He even engineered database components to outperform their OS complements. "The largest customers need scalability and the largest HP boxes," he added. "If you work in the low-end and midsize end of the market, you will hardly find an HP-UX box anymore."
The current state of HP's Unix business will be reflected in its details on Integrity servers and the Business Critical Systems numbers. This is about the only metric we follow closely; ProLiant sales for Windows and Linux (the ISS group) have been healthy. BCS includes OpenVMS and NonStop sales, but HP-UX is a very large share, too. The future for these environments relies on Itanium and any new customers HP can cobble up for its Unix.
Oracle Harm to 3000s & HP, Past and Present
Oracle has become a loud competitor to HP in the migration alternative game. HP took an hour's webcast this month to assert that the Oracle-Sun-Solaris action is mostly heat without much light to lead the way. But the database has certainly gained HP's attention now that ousted HP CEO Mark Hurd is leading the Sun attack. If some stories are to be believed, however, the current fracas is a long way removed from Oracle's blows against the HP 3000's futures.
At one point HP was eager to keep its 3000 customers buying Hewlett-Packard enterprise solutions. And by 2001, the story goes, Oracle wanted a clear path to selling Oracle hosted on HP's Unix servers. HP was going to cut something out of its merged product line with Compaq. The 9000s were never on the block, but there was the HP 3000, sporting tens of thousands of systems, nearly all of them running an IMAGE SQL database.
So here goes: HP didn't kill the HP 3000, Oracle did. Oracle made HP a deal they couldn't refuse. Stop selling competing database software, and Oracle would partner to sell HP-UX systems as Oracle servers. Since MPE/iX is tightly coupled to IMAGE/SQL, this translated to the end of the HP 3000. The smoking gun was supposedly this: As part of HP's announcement on discontinuing the HP 3000, it included the end-of-life for Allbase/SQL on both MPE and HP-UX.
The theory has some credibility. In 2001 there was a lot of growth potential for Oracle-plus-HP-UX. Oracle had grown up plenty in the decade before that fateful date (when its CEO Larry Ellison, left, was a tender 47 years old.) In 1991 his juggernaut was pretty much out of the game, even with 20,000 customers, because it was scraping the bottom of its cash barrel. Some reports said Oracle creditors were ready to call in their 1991 notes. Those 20,000 sites arrived in the fold because Oracle got ruthless about sales and promises and vaporware delivery. $80 million in cash from Nippon Steel in exchange for a piece of the Oracle Japan arm helped Oracle back from the brink, too. Then there was a forced restatement of sales all the way back to the company's first quarter as a public company. Booking nonexistent products and stretching future contracts as current revenues, it was all the kind of behavior old-school sales reps would chuckle about today -- a day when Oracle wants to rock the 2011 boat, charging that HP is paying Intel to prop up the Itanium product line.The old-school files here in our office include InformationWeek and Computerworld stories about how Oracle came back from the brink in 1992-93. Oracle users frustrated by wait for DBMS, tools said a story by Jean Bozman. She was a reporter in those days before becoming an IDG analyst, one who figured the world still had more than 25,000 active 3000 systems running, two full years after HP dropped its axe on its 3000 unit.
Over at InformationWeek, the "hungry to profits" company was struggling to deliver Oracle 7 (the latest version is 11, with 12 in the wings). A list of that year's Oracle Business Alliance Program members didn't include HP, even though by then Hewlett-Packard's strategy was in total thrall with Unix. Oracle still was the leader in Unix databases in that era, selling just a bit more than half of Unix RDBMS installations. Sybase was jousting at Oracle with a preannoucement of Version 10 of its database, which one newsletter said was "an announcement, if it's possible, with even more hot air than Oracle's Version 7."
Meanwhile, there was no hot air in the IMAGE of that period. HP was being led by the nose by its customers to keep IMAGE an integrated part of the 3000 solution. Hewlett-Packard wanted to separate IMAGE from the 3000 and make the database an add-on. A revolt at the Interex user group conference of 1990 ensued. Even though the users carried the day -- and Jim Sartain became a business-savvy IMAGE R&D manager afterward -- the IMAGE standard bearers look see this as the start of the decline of HP's attention to IMAGE, business-wise.
Oracle couldn't be bothered with current 3000 R&D from that point onward. Jennie Hou was yoked to the HP's thankless task of getting Oracle resources devoted to MPE/iX -- because HP hoped Oracle would attract customers in the 3000 arena. Oracle just didn't eager to attract any off of the 3000 platform, maybe because developing for MPE wasn't the Unix business Oracle had used to get off the mat. Or maybe Oracle saw the folly -- which it first told me in 1985 -- of competing with a bundled solution like IMAGE. While the rest of the industry was deploying Oracle 8, that release was out of reach for the 3000 user. We wrote in 1997:
It only took one Oracle sales rep in California to get the 3000 customer base worried about the future of the database on MPE/iX. One rep's comment to a 3000 customer circulated through the Internet, asserting that Oracle was only going to support its database through version 7.2.3 for the HP 3000. This led to a fair bit of piling on, as people wondered what the purported pullout meant for the HP 3000 and why anybody would want to get serious about using Oracle instead of IMAGE anyway.
HP and Oracle went to work on damage control almost immediately. The two allies whipped up a quick update on their plans for the HP 3000. The sale's rep's comments were based on a partial truth: Oracle is still not willing to commit to a list of supported platforms for Release 8. Despite what some might see in the tea leaves of whether the 3000 is mentioned on Oracle's Web page roll call, no one using any platform knows for certain when they're getting an Oracle 8 -- not just yet.
Oracle 8 on the HP 3000? One year later, HP had committed its own engineers to just getting a fresher Oracle 7 onto the platform.
CSY has engineers working on the port of Oracle 7.3.4 according to Jennie Hou, the manager of the HP 3000-Oracle relationship. “The porting resources are still engaged in 7.3.4,” she said, which HP expects to be available to 3000 customers within calendar 1998. Version 126.96.36.199 is currently shipping for both MPE/iX 5.0 and 5.5 HP 3000s. There has been no announcement of an Oracle 8 port from CSY yet, “because customer needs are being met by Oracle 7,” Hou said. Oracle-based applications are ready for the HP 3000 in manufacturing, financials and human resources, and Oracle plays a part in data warehousing solutions for MPE/iX.
An Allbase pullout as a smoking gun would be hard to point at HP's 3000 history. The database had its fans among some 3000 sites, but it seemed to be more of a "other option" item on the HP pricelist. More than 95 percent of the 3000's sites were running IMAGE/SQL and still do. That's one reason that Eloquence has done so well as a database migration replacement for 3000 sites: it behaves just as IMAGE does with 3000 programs that are moved to HP-UX or Windows platforms.
3000 users watched a lot of one-sided pursuit of Oracle affections during that decade leading to the pullout. By the fall of '98 it was obvious HP 3000 customers didn't want to fork over tens to hundreds of thousands of dollars for a database instead of IMAGE. Winston Prather announced the end of Oracle futures on MPE/iX at the HP World conference -- one year before taking over as 3000 GM which led to, well, the end of HP's futures altogether for the 3000.
“The current plan is that there is no plan to port Oracle8,” Prather said at the 1998 forums. “The real fundamental issue: Oracle’s position is that, ‘We’ve worked really hard with you to bring it to the platform and done joint marketing programs, and it’s not working.’ Oracle is looking at it and going, ‘Why am I doing this?’ They haven’t come close to recouping their investment. Their current feeling right now is that there’s not enough business on the 3000.”
Customers using Oracle say the database runs faster on HP 3000s than on HP 9000s, something that HP might want to use in making a case for additional development resources. Prather noted that customer adoption has been slight in the face of a more cost-effective 3000 database: IMAGE/SQL.
“It’s very expensive to implement an Oracle solution relative to a TurboIMAGE solution or even an Allbase solution,” Prather said. “The bottom line is that 3000 customers like TurboIMAGE. They don’t want to leave TurboIMAGE. Our application providers don’t want to move to Oracle, because they like TurboIMAGE."
It's not like HP wasn't trying to sell the database. In 1996 it offered Oracle's 7.2.3 version to 3000 sites at prices starting under $1,200 per seat with an eight-seat minimum. For the first time, you could get a small 3000 into Oracle for just under 10,000. The price per-seat increased based on HP's CPU tiers of the day -- that under-$1,200 price was only available for the lowest 3000 tier. Price was a barrier vs. an included database, and then there was migration.
The HP pricing doesn't alleviate the roadblocks of data migration and increased management demands. Taurus Software's Forklift migration tool, which uses a graphical interface to map IMAGE datasets to Oracle tables by way of the Taurus Warehouse utility, was the start of moving data as easily as any tool on other platforms. Forklift gave managers a visual aid to get IMAGE data into Oracle databases.
Within five years Oracle wouldn't have IMAGE to dodge around anymore while it sold HP's systems. Wht good did that do in the long run? By this month, Oracle doesn't even want those HP Unix systems to exist. Its charge of paying Intel to keep Itanium alive is pretty blustery with hot-air, even by Oracle's standards. HP doesn't make Itanium anymore, even though its engineers retain a role in processor design. Itanium is an Intel product by now, and if HP is chipping in extra dollars to keep development going, that's in HP's best interest to keep selling its Unix servers.
We find it interesting to see how Oracle has crept back from the "Itanium is dead, and Intel isn't saying" hot air of this spring. Intel came back with an opposing gust that might have knocked long-time yachtsman Larry Ellison off the "attack Intel" tack that's part of his warpath on HP. Oracle can't hurt the 3000 anymore, can it? That depends on how you think of the Integrity servers as 3000 migration replacements. This Oracle war is creating distractions for HP's Integrity sales.
November 18, 2011
Last Words from First Users on HP's Pullout
All this week we've been marking a tenth anniversary of HP's ill-fated decision to pull out of the 3000 community. There have been other things happening besides the remembrances. But there's little happening in the community today that has not been altered -- for better or worse -- by the Hewlett-Packard choice. We also have a package of pullout stories coming in our November print issue, along with photos from the community's first HP3000 Reunion. But we'll wrap up our Pullout Week with stories from two key community members. Jeff Kell started and maintains the HP3000-L mailing list at utc.edu, where 3000 discussions and tech tips started in the early 90s -- and remain online today. Kell was also a SIG leader while volunteering for the Interex user group.
Then there's John Wolff, an initial board member of OpenMPE who first joined HP in 1968, and then became an HP customer in 1974, and started using the 3000 in the system's Classic days -- and so has felt some of the deepest disappointment. But he still watches the company for signs of hope.
Jeff Kell: As of the mid-1990s, essentially all of the University of Tennessee at Chattanooga's business applications were all legacy applications on the HP 3000, having evolved from the initial roots of the student/admissions/grades/records system developed in the mid-to-late 1970s. One was a third-party Library application added in the 1980s, but still HP 3000-based. At our peak, we hosted five production HP 3000s in our server room covering administrative, academic, and library services.
Academic usage migrated first to IBM, and later Sun-Solaris/Unix, but business applications remained intact. Traditional "internet" applications (e-mail, file transfers, Gopher and later WWW, etc) grew on Solaris and later Linux.
An initial investigation into a third-party student system led to an attempted "migration" in 1997, based on a large-ish HP-9000 quad-processor system with a sizeable disk array. Dissatisfaction with the software (relative to the 3000 legacy applications) led to a delay in implementation of all but the student financial aid and accounts receivable systems. At that time we began to "fortify the foundation" of the long-term viability of the 3000 platform. We were well into MPE/iX and the Posix environment, and there appeared to be some real solidarity given these capabilities (the lack of "Internet readiness" was often used to criticize the platform).
The 2001 announcement was a knife in the back of our long-term planning and objectives, from which we never fully recovered. The original Library application (3000-based) was moved to Linux/Oracle (where it remains to date). The partial third-party student implementation on the HP 9000 was moved to Linux/Oracle -- where it too remains to date.
Parts of our identity management system, as well as some percentage of student records which did not survive the automated migration, remain on our HP3000; but the system is essentially running "read-only" as of this year.
We do still have a number of HP's printers. But we have never since seriously considered them as a business, instructional, or even personal computing platform anymore. Caveat emptor.
John Wolff: My HP Systems Engineer at Laaco, Ltd. was visiting us a couple of weeks before the official announcement and gave some strong hints about what was coming. So the actual announcement was not so much a shock, but rather a validation of a great disappointment.
In my opinion, 2001 was a watershed year for HP, as it began a lost decade of bad management and poor decisions. The company is still struggling with a bad Board of Directors and the seemingly endless consequences that flow from that. The agonizing studies and public review of strategic questions over a period of months, like the Personal Systems Group spin-off and the TouchPad/webOS debacle, illustrate this far better than anything I could ever say. There is nothing more destructive to a business model for employees, customers and suppliers than failures of decisiveness, of commitment and exectuion.
I began my career with HP straight out of college in 1968, when HP was widely recognized as one of the best managed companies in America. Imagine how it was to transition from a proud six-year employee into a satisfied customer for 30 years. I felt like I knew a secret: That HP was a terrific vendor with great products and strong support that was making my efforts on behalf of our company a success.
My company was primarily in the business of owning and operating private clubs when I started with Laaco in 1974. We developed a custom club system on HP 9830s, which we used until 1986. Beginning in 1982 we started developing a new system on a Classic HP 3000/44 and started using it for production some 25 years ago. Our custom application continued to grow with continuous enhancements over the years, while the hardware was upgraded seamlessly to a Series 48, Series 58, Series 70 and finally to a PA-RISC Series 928.
Meanwhile, we reduced our exposure in the club industry from four clubs down to two as the company began moving into a different industry, self storage. Although we still have the two remaining clubs, there is little growth in that business, so we did not have to expand to faster hardware. But we did continue with our custom development, which is primarily written in Transact. I believe we hold the record (by far) for the longest use of the same platform in the private club industry, where it is typical to switch to a new system every five years, if not sooner.
Now, as I mark 37 years with our company and assess our club system strategically in relation to our corporate direction and a dominant role in the self storage sector, I find that it is time to make plans for the future. My programmer is almost 72 years old and has been with us for 29 years (another record). It does not seem realistic to go looking for another Transact programmer within the shrinking HP 3000 ecosystem. Consequently and with reluctance, we have begun evaluating a replacement system from the traditional club software offerings that run on Windows. This conversion will probably take place next summer and demote the HP 3000 to archival duty.
Finally, with the benefit of hindsight, I must say that selecting the HP 3000 30 years ago was a great decision that paid off as both a development and production platform, in spite of recent HP mistakes. I have no regrets regarding the decisions that I had control over; I can only wish that those decisions beyond my control could have been otherwise.
In 2001, I began to watch this once-great company start a decline over a period of 10 years into one of the worst-managed companies in America. I am left to wonder when HP will hit bottom and recover its sense of identity and direction. We all continue to watch hopefully.
November 17, 2011
Some couldn't believe the pullout at first
Some of the members of the 3000 community had no reason to believe HP would pull out of the 3000 business. In this week that marks the 10th anniversary of that exit, community members are sharing their stories of where they were when they heard -- how much they felt they could believe -- and what's become of their careers since then.
Brian Edminster: I was subcontracting at a company that specializes in supporting medical information systems (primarily Amisys, but others as well). This was a new contract at the time, and came after a multi-year gig doing a Y2K conversion on a large legacy Retail Management system.
I almost didn’t believe the news — there were too many other big changes happening in the world — and HP management had recently redoubled their support of the platform, so I just couldn’t believe it at first. I guess I was still expecting the New HP to act like the Old HP.
My consulting practice has been stable, with slow but steady growth. I’d say that my career has taken directions that I’d not have been able to anticipate, just a few years before. I’d not have gotten into open source software on the platform if the ecosystem of commercial software hadn’t started drying up. I wouldn’t have been able to justify going to the last Greater Houston RUG meeting to present a paper, and I wouldn’t have started building a website to act as a central repository for free and open source software for the 3000 (www.MPE-OpenSource.org).
Robert Holtz: I was working away on the COBOL and FORTRAN programs that were the heart of the Computer-Aided Dispatch and Mobile Data Terminal programming that ran on the three HP N-Class 3000's our Phoenix Police Department had upgraded to -- just earlier that year.
Christian Lheureux: I was sitting at my desk at Appic, sorting through HP3000-L newsgroup postings. I learned that the HP 3000 was going to be terminated from an HP internal source that I could absolutely not quote due to an ongoing NDA. In fact, I had been informally tipped much, much earlier that this was going to happen, but I simply could not believe it !
If you were there 10 years ago, you probably remember that some emotions ran quite wild, and that certainly includes mine. After a while (weeks, months), I remember having a big sigh and realizing that, in the aftermath of the Compaq takeover, HP would not keep 2 proprietary platforms and that, between a 71,000-unit installed base (HP 3000) and a 700,000-plus-unit installed base (VMS), the choice was quite obvious. To this day, VMS still exists. They even recently introduced a new release.
The company I worked for that the time still exists as a software publisher. We went bankrupt in late 2005 and the company was finally liquidated 1 year later. I probably sold the last four new HP 3000s in France, on Oct. 31st, 2003. I did my last significant MPE assignments in 2004. After that, my HP3000/MPE activity rapidly became marginal. When our company went bankrupt, I was immediately made redundant. Therefore I have absolutely no idea of what happened to the systems I had in the datacenter -- well computer room. They probably ended up in a garbage dump, much like an unneeded refrigerator that burns too much energy.
I later did some HP-UX work, then became a sales exec, then went back to pre-sales, which I still do today. I've been part and parcel of the HP ecosystem for all my adult life, HP user as a student, then HP employee, then HP consultant, then HP partner. My HP-UX skill level never rivaled my MPE knowledge, not even close, not even by a long shot. And, perhaps more important, the fun I had doing HP-UX stuff never came close to the fun I had doing MPE things like debug/dumpreading, executable code troubleshooting, performance measurement, developing tools that I needed for other assignments, writing stuff, "educating" customers, etc.
There used to be two documents that I wrote on the OpenMPE website while I served on the board. One was the DAT compatibility matrix, and the second one was the HP(e)3000 line-up, sorted by software tier, complete with performance indicators. I have absolutely no idea whether those documents still exist and if they are available anywhere. My best guess would be that,10 years after, no one cares.
That's history being made. Things come and go.
John K., AOL: I was sitting at my desk in AOL's Reston Technical Center in Reston, VA, when I heard the news. I was the manager of the Access Wardialer Lab, which filled a little over 1,100 sq. ft. of raised floor with racks containing hundreds of test and measurement PCs connected to three DS-3 lines providing telephone lines.
We had one HP 3000, and it collected, stored, and analyzed access wardialer data from hundreds of PCs which called every AOL dialup number multiple times every night to test the dialup network and hammer the AOL Windows client. The HP 3000 produced a number of reports, charts and emails every day, with virtually all of AOL's senior executives and management on the distribution lists of those emails. It also hosted a web site for retrieval of reports, processed wardialer data, Windows "debugview" logs, and other analytics. I'm told that the HP 3000 was turned off and stored for somewhere between 12 and 18 months, and then converted to an HP 9000 (AOL had many, many, many HP 9000s).
AOL's dialup usage took a nose dive in 2004, and in late 2004, my group was disbanded (layoff). Since then, AOL has split off from Time Warner. The AOL Reston Technical Center where I worked no longer exists. I was invited to, and attended AOL's 25th Anniversary Celebration in Dulles, Virginia, on May 24th, 2010, and it was great seeing so many of my former co-workers, most of whom have moved on to other jobs in the various tech industries. While a manager at AOL, I also coded in SPLASH, SPL, BASIC, and BUSINESS BASIC, and I created both terminal-based and web-based applications.
Today I'm the Software Engineering Manager for an Internet Services Provider which also provides hosting, co-lo, and VoIP telephone services. I still code, but now I code primarily in PHP and SQL, and the company's Enterprise Information System (EIS) is of my design. I also wrote all of EIS's core code.
November 16, 2011
Things change, some 3000s remain the same
When we polled more than 30 customers of the HP 3000, we were surprised how many still employed their systems a decade after HP left the field. Some are using the same servers which ran on the day HP predicted the demise of the ecosystem, 10 years ago this week. Others have relegated their systems to archival duty. We heard from a few that've turned off 3000s completely since 2001.
At the Catawba Valley Medical Center in Hickory, NC, Jim Dellinger said the center's 3000 has been decommissioned quite awhile. "The HP 3000 was discontinued here in 2004," he said, "and hardware services moved from LAB (my niche) to IT. I'm sure there are no HP 3000 servers there."
Dave Powell, whose report on his 2001-2011 3000 experience appears in our November print issue, told the world more than six weeks ago that his company in the fabrics industry is moving off the 3000. "MMfab has decided to migrate," he said. "Buy a (gasp) package. Toss the system I've been working on for 30 years." But still run a 3000 in archive more for the next 1-3 years, once the real implementation work starts at MMfab -- and gets completed.
But for every report of a departed 3000, we heard two that were remaining on duty. At least for the next several years. Connie Sellitto, who had about two weeks to solve the problem of "wireframing" her 3000's app architecture for a migration in March, checked back in to say it will be several more months until anything based on .NET is running the US Cat Fanciers Association.
"I was working as Programmer/Analyst at the US Cat Fanciers' Association (CFA) on our third HP 3000, a Series 937 RX, when HP announced its end-of-life," she said. "This really scared a lot of people, but I kept telling them we had third party hardware and software support, and not to worry. The company directors at the time decided to leverage the 350-plus programs with a migration to an HP 9000 -- and we in fact secured a used system, only to have them reverse their decision and opt instead for a newer A400 3000.
"The new HP 3000 remained in use in the New Jersey office until July of this year, at which time it was transported intact to CFA's new location in Alliance, Ohio. Plans were to (really, this time) migrate off the MPE platform entirely, with a complete rewrite on a .NET SQL-based system. This project, originally underestimated to take 3 months, is still in the development stages, and although I've moved to a new job, the 3000 is still going strong. It continues to run the business-critical operations for CFA."
"And the estimated time to finally shut down this venerable 'legacy' system? My personal guesstimate is another 3-4 months. My only desire is that the data be secure and that all business practices be enabled. Long live the HP 3000!"
Peter Eggars: I was off with friends celebrating my 49th birthday. By that day I had too much time, too much money, had lost much of my obsession with computer technology, and lost my faith in HP. I had been told over a year before that it was coming, and didn't hear about the official announcement until much later. It wasn't until I had a long afternoon discussion with Wirt Atmar that I comprehended the importance of the day, and the missed opportunity to have done anything about it.
In hindsight, the spirit that allowed the HP 3000 to grow and thrive in an IT environment that was dominated by IBM (it has 80 percent of both hardware and software market shares) was lost with the embrace of the new IBM, as well as Microsoft, who were able to take shady business practices that monopolize a market to new lows.
The HP 3000/MPE could have evolved into the premiere Rapid Application Development platform for small- to enterprise-class business applications using a Linux kernel and drivers, one of the GUIs, and an open source database. Integration with Open Office (now Libre), would have been icing on the cake. I think there was a good chance HP could have beaten Oracle, had HP started down that track in the late 1990s. But I have to admit now that Wirt Atmar was right -- 2001 was the last possible year that could have been successful.
Gilles Schipper: Unfortunately, I can’t recall where I was. I do remember first hearing an inkling of it from Wirt — and, coming from him, even though it was not confirmed, I knew it would turn out to be true, as of course it was.
Despite all these 10 years that have passed, my company GSA still has enough customers that I perform HP 3000 System Administration and support duties, staying reasonably occupied and earning a living -— although, of course, not to the same degree as 10 years ago.
What I miss a lot is the annual Interex conventions that afforded the means to revisit with old friends such a you. Even a few years back, a couple of GHRUG meetings in Houston were terrific. Unfortunately, I was unable to attend the recent Mountain View get-together at the Computer Museum. Hopefully another opportunity or excuse for another conference/meeting/get-together will arise in the future.
November 15, 2011
Now in an 11th year of post-HP: user reports
We're continuing with the community's first-person testifying about HP's November 14 pullout from the 3000 market 10 years ago. Today is the first day of the 11th year of the rest of your life, because HP's never going to go back on its decision to cease making, enhancing or, in most cases, supporting the HP 3000.
But we've heard from users who hoped otherwise. Many did in the first few years after 2001, because it was hard to believe from the beginning. At least difficult for users and suppliers who knew so many satisfied 3000 owners, or were making a good living off an ecosystem HP proclaimed as mortally wounded.
Why look backward at an event nobody will ever change or recant? You can get hope from the new ground which some of the users have attained. And you'll see how to manage such a sudden change of strategic direction from a supplier, though some of these stories. Plus you can believe that it can happen to any product controlled by a single-vendor. We asked: 1. Where were you when you heard the news, and what became of the 3000 you were using, and 2. What's become of your career and company over the last 10 years.
Bill Towe: I remember attending the HP World shows for 1999 and 2000 when HP announced it was opening its arms to the HP 3000 and would continue the line, and the future seemed safe. Then barely a year later, I was attending an HP Channel Partner conference in Las Vegas when I heard a rumor that the HP 3000 was back on the chopping block. I couldn’t believe it, because only months before, CEO Carly Fiorina had informed the HP 3000 collective that we would see the MPE systems line for years to come.
During that Conference, I learned the HP 3000 was finished and would start a phase-out of equipment process followed by the End-of-support death march. I was simply shocked. My company, BlueLine Services, was only two years old at the time and 95 percent of our business was MPE system sales and support. We spent the next few years holding out hope that HP would continue to postpone or completely reverse their decision to end the HP3000 line.
"Over the years, it has become more and more difficult to be an HP-Only reseller. Since that fateful day, we have become an HP, IBM, Dell, Compellent, Cisco, VMWare and HDS reseller, as well as provider of managed services and cloud Computing Services, coupled with hardware and software support for MPE, HP-UX, and Windows OS. Since the dissolution of the HP 3000, my company has diversified to the point that HP no longer has the lion’s share of what we provide our customers. I still find it difficult to believe that the same manufacturer that created the greatest hardware and software system ever produced, also ended it and so unceremoniously. Sad."
Chris Bartram, 3k.com: When I heard, I was working a long-term consulting contract managing HP 3000s and several datacenters for the US government. My company 3k Associates still exists and its HP 3000s are still humming, although only one of threee stays powered on these days.
My job that pays the bills these days has nothing to do with HP 3000s -- and thankfully, very little to do with HP at all.
Craig Lalley: I was working from home for Lund Performance Solutions at the time. The demise of the HP3000 was greatly overshadowed by the events of Sept 11th, for me. Sept 11th had a huge impact on the economy, as well as my personal economics.
I guess was expecting HP’s decision. HP’s actions were much louder than HP words. I believe HP had decided to end the HP 3000 several years before. The sad part is they could not admit it. I still have a love for MPE. I believed there still was a place for a proprietary OS in the business marketplace. Sadly HP did not feel the same way.
My biggest disappointment is the loss of the HP 3000 user groups and the community they inspired. Sadder still is the loss of some major community members. I must add that I enjoy the HP 3000 “e-community” of friends I have met and worked with over the years. The community still consists of some very special and highly talented people.
My greatest hope for future of MPE is not OpenMPE anymore, but Stromasys and the HP 3000 emulator, Charon HPA/3000. To date, I know of only two major bugs/issues, and performance is the next objective. I hope the hidden HP 3000 homesteaders will find a way to see a demo of Charon HPA-3000, I am sure they won’t be disappointed.
November 14, 2011
One Decade Later, You Survived HP's Pullout
So here we are, 10 years to the day since HP announced it would not be participating in the future of the HP 3000 anymore. It's still Nov. 14 in most places where the NewsWire's blog is being read, so here's a history that nobody's been dying to tell. It's about customers of a company that thought it was killing off a computer line. Instead, it killed off a lot of its customers, in the sense that thousands of them though HP was dead on arrival in the IT futures department after Nov. 14. There were careers and companies killed, too. But we've never called it the End of Life for the HP 3000. It's always been the end of HP's life with the 3000. By this year I think of it as a pullout, an act that signals a loss of will and faith, forgetting why you got into a relationship in the first place. HP had given up on a group that I was glad to dub "homesteaders." Everybody was one, until they could manage a migration, after all.
The shock and outrage on that Black Wednesday was astounding at the time, because HP didn't whisper a word about the customers who could never leave a efficient, vital platform. The first HP message was filled with warm concern about getting everyone onto the right computer as soon as possible. As if the bolts and boards of the 25,000 systems working worldwide were about to go toxic or something. HP couldn't even pin down when the closing date would for the 3000 division, CSY, then headed by Winston Prather.
From a CSY perspective and a support perspective, it’s business as usual for the next two years. It’s time for customers start their planning to move to a platform that will serve their businesses better in the future. HP recommends that customers begin transitioning off the HP 3000 to alternate HP platforms.
Customers were not surprised at the news, Prather said, "and they really appreciated HP being able to tell them what we see as the future role of the platform." Prather said these top-tier customers of late 2001 "already have a multi-OS strategy, so they’ve been evolving their applications over time. It is a stake in the ground, but the CIOs I talked to were appreciative of hearing what the future holds."
As proven by the reactions of the next two years, Prather had only talked to companies who could afford to migrate -- and were grateful for an infusion of truth from HP after years of everything else.
For the last week I've been asking the community members to tell about where they were on that star-crossed day that they heard the news — plus what's become of their careers, companies and the computers running on Nov. 14. We've got a massive feature coming in our November print edition, which went to press at 5:30 this morning.
The outpouring of memories and updates and resolve for the future has been profound and prolific. But this has always been a group that knew how to say what they meant, and how they felt. "I'd say we've all been a pretty good human chain holding the 3000 Community together," said Jack Connor, who just departed the OpenMPE board. He was kind enough to note that the stories and articles here "do a lot to make us aware that there's indeed life after HP, and a pretty full one so far." Considering the promise of an emulator and the state of virtualization today, the last decade could have unspooled a better future than what HP delivered.Some of us had a little advance notice, and some said they saw the move coming before HP announced its exit. My preview involved a phone booth, a cassette recorder with a suction-cup phone pickup, and an Internet Cafe. That's how long ago this all happened. Somehow to a lot of you, it feels like it was only a moment ago.
We leave the first word in the storytelling here on our blog to Alan Yeo, who commissioned a pair of iconic editorial cartoons to express the feelings of customers who didn't get a heads-up, but felt blindsided. Some were even resellers, who learned about the pullout at the same time as their customers.
Yeo wrote us this morning about his "second Black Wednesday," noting a British event that seems as much a cock-up as dropping a profitable platform and its customers at great loss to the little guy. He advised we read up on Wikipedia about that other Wednesday in the UK when £27 billion was spent while a speculator made $1 billion.
On the "Second Black Wednesday" of Nov. 14, I was waiting for America to come on-line and find out if what I had been told a week earlier was coming was true or not. Unfortunately it was.
What became of the HP 3000 we were using at the time? It's still running and still used for development. When you're developing migration software to replicate HP 3000 software, there is no better place to test than under MPE.
My strange observation is how different the last 10 years could have been, if HP had acted differently. Sitting on my desk this week, 10 years after HP announced the end, is a little blue box with a silver-blocked Stomasys logo on it. Inside is a USB stick that will hopefully this week allow a brand-new Intel i7 server that is just being installed here to boot up as an HP 3000 running MPE.
Ten years on, and there is still a potential market for an emulator? We have had a decade where it appears at every turn HP took the wrong path.
Whilst a few people may have been disappointed, most people would have slowly moved and been more than happy if: Instead of announcing the EOL for the HP 3000 — because upgrading MPE to IA64 was too expensive — HP had decided that they would support MPE running in 32-bit on Itanium, or even in a VM under HP-UX on Itanium. Remember even today, most "modern" server applications are still running in 32-bit mode on 64-bit architecture. HP might have sold quite a few more Itanium Servers and certainly would have kept a lot more customers. And there are a lot of vendors around the world that would have had a completely different decade.
Even though they decided to EOL the HP 3000, it then appeared they did their best to frustrate the third-party ecosystem which they indicated was expected to carry the load. The couple of extensions to their end of support dates, well, those damaged both the third party support vendors and the migration vendors — as the need for users to make any decisions was pushed out. Which is probably why there may be enough users still around to make an emulator viable. The Chinese curse goes, "May you live in Interesting Times." And the last decade has certainly been interesting.
November 11, 2011
Take a search into our print issues, online
Every three months, The 3000 NewsWire returns to its roots and arrives in paper and an envelope. It's a format that we first delivered 16 years ago. No matter how fast the Web can update us, there's a different pace of education and insight that comes off the printed page. We're fortunate to have the support of community sponsors to keep our print issue alive. (Frankly, it would be hard to get me to stop creating it, coming from a "stop the presses" heritage at newspapers here in Texas.)
But onward to my point: You can read those printed issues here on your PC or laptop, created as PDF files. The latest one is always available from a click at the top left corner of our main website page on this blog. We also have an archive which is summarized below, with links for downloads.
This month marks the start of our 17th Volume, as we old-timers say in the publishing business, each of those volumes one year of printed editions. Our volumes begin in the fall, when the 3000 NewsWire was born. As it turns out, November became a pretty important month for the 3000 -- but that's a story for next Monday, Nov. 14.
One advantage of creating a PDF file of these print issues is that they can be searched with a PDF reader. All of these articles are included in our blog, after selected stories get "Print Exclusive" status -- first available only in paper. But if you want to enjoy the print presentation -- a magazine style, if you will -- click below. If you'd like to have a copy arrive in your postal mailbox at the soonest date, send an email to me at the NewsWire and I'll ensure you're on the mailing list for premium paper. Be sure to include a mailing address.
So we'll see you in the papers -- you never can be sure whose picture or name was in the news on a page that's printed. It's worth a look; collect 'em all, as they used to say in the comic books. Maybe there's a place for a Ken-Do revival in 2012's printed NewsWires.
November 10, 2011
HP puts down Oracle, which puts up Solaris
Hewlett-Packard summoned a market analyst to tell its HP-UX customers "Unix is not going away in the near future," said Dan Olds, founder of Gabriel Consulting. "Probably not even past the near future."
These kinds of assurances are needed in a marketplace where Linux has all the buzz and Windows all the populace among environment choices. Sure, the apps drive the OS choices, but a company's got to ensure it can retain and train the IT pros who keep apps alive and moving, instead of stalled. As one example, the replacement-for-IMAGE database Eloquence keeps growing "to keep your applications from stalling," said founder Michael Marxmeier.
Eloquence runs on Windows, Linux and Unix. It the last arena it becomes a player in the ongoing saga of What Will HP Do About Oracle? Oracle sells one of the biggest competitors to HP-UX, Solaris. Last week HP said that Oracle's Unix, Solaris, is a distant third in customer support and deployment to HP's Unix. This week Oracle announced a Version 11 of Solaris, which will be marketed as a cloud-friendly OS. It's also got tighter integration with Oracle's database, a strategy HP once used to great effect in the HP 3000, with MPE plus IMAGE. Here's a telling passage from a Computerworld story on how the link-up works for Solaris: "By controlling an entire stack of software, the company can make holistic decisions over which part of the stack would be best suited to tweak to gain performance improvements."
That's not in the field to be surveyed today, and perhaps not even this year. The message during the one-hour HP webcast relied upon a 2010-11 Gabriel survey of companies already using some kind of Unix. Prompted by HP's Katie Curtin-Mestre, Olds said that Oracle is well behind HP and IBM in categories like OS Quality, Patching Quality, availability as observed by a user base, plus a lot more. Olds -- whose 15 years in IT includes nine years of marketing consulting for hire -- said it's a two-vendor race for commercial Unix, where "Oracle is not as competitive as they would hope, or we expected. In some areas, since Oracle took over Sun and stabilized it, they've lost some ground."
HP would be happy to learn that continues to be true. The Gabriel survey was taken during the first six months of work by Mark Hurd, former CEO, who has led the Sun/Oracle rebound culminating in a fresh Solaris and refresh of the SPARC chips. How this webcast chest-thumping by HP will imact its wish to get Oracle to love Itanium/Integrity once again -- well, that's anybody's guess. But it's hard to portray the webcast as an olive branch.
One thing feels certain: when you use red as the color to depict a vendor on a PowerPoint slide, it's never a friendly label. HP's customers are observing that Solaris is much slower than HP-UX, but HP's Unix is just 1.3 percent faster than IBM's Unix. This may be a two-horse race in observations and speed. But HP needs Oracle to keep its customers on HP-UX servers, and coloring the company red looks as combative as suing its rival to keep supporting HP's Unix.Other high points of the HP-Gabriel survey (Olds calls the study a Vendor Face-Off):
"It will be interesting to see what happens, now that Oracle has released some new systems. We'll see if they resonate with customers." (We're not certain how Oracle will participate in the new Face-Off, but the vendors don't need to be part of the polling. Some vendors do pay for webcasts of the results, though.)
This kind of report is no new tactic in getting customer loyalty pumped up. The objectivity comes under scrutiny when an outside consultant says, "It's clear that the commercial Unix market is primarily a two-horse race between IBM and HP. Oracle hasn't managed to turn Sun around in any significant way."
Olds said, "Customers are still using more Unix capacity than they used the year before. There's a lot of churn, as they move stuff to x86 and take out Unix platforms. But they're also adding Unix platforms. The amount of capacity shift is more than what's being turned off." So the Unix advocate is now looking at capacity rather of footprint. They've lost the war on server counts, although HP continues to report its Unix servers lead that sector in market share.
"The [capacity] growth isn't as fast as it was back in the day of the early-to-mid '90s," Olds said. That's the period where HP is plundering the HP 3000 base for converts to HP-UX, and Windows doesn't have any traction as an enterprise environment. Linux was little more than an experiment back in the day. Olds then asserts that "nothing's growing like it was back then." That will come as news to Red Hat, SUSE and the other Linux providers.
"Unix has moved to a different place; it's become that mission-critical platform," Olds said. Right on the money, from the reports of the former 3000 sites who've turned to Unix. None of them report that Unix is any more mission-critical than the 3000, but we don't have a superb window into the largest of Unix customers. For the most part, that size of site didn't use the 3000.
Fewer shipments, larger systems, greater uptime needs: Some of this sounds like the HP 3000 situation of the late 1990s, when footprints of MPE/iX were being overtaken by Windows and Unix. Nobody knows the future for sure, even when a vendor tells the tale: witness the assurances of HP World 2001 about the 3000 vs. the pullout just three months later. But the future of Unix is now tied firmly to big systems and mission-critical capacity. It's hard to see how that competes with the nimble and smaller options from the cloud, even if HP says Unix is a part of its cloud computing plans.
HP's Curtin-Mestre said "we get a lot of questions about Integrity [servers] and Itanium, and to add to your point, the Itanium market revenue is larger than AMD as a whole." That's a lot of sales, but now the vendor is comparing its Unix business to a smaller subset of the Windows x86 alternative. "This whole conversation reminds me of the old adage from Mark Twain, that 'rumors of my demise are greatly exaggerated.' "
It was a "report" rather than rumors that Twain replied to, for those who admire the humorist's accuracy. He lived another 13 years after his handwritten note to Frank Bliss. The note sprang from reports Twain had died, but it was his cousin who'd been ill. People like to think of operating environments as having a lifespan, especially HP, which encouraged the thought that MPE/iX was dying, along with the HP 3000 and for that matter, IBM's AS/400. We'd rather consider Twain's writing about death in 1906 in Mark Twain in Eruption as a means of getting the most honesty from the departed.
I think we never become really and genuinely our entire and honest selves until we are dead--and not then until we have been dead years and years. People ought to start dead, and they would be honest so much earlier.
Perhaps as the MPE/iX OS moves into its 11th year of being "dead," HP-UX can hope for a similar future. If nothing else, the vendor won't need to hire outside analysts to prove people are still using and choosing its proprietary OS.
November 09, 2011
You've got the power, but you can use less
Last week we we tried to discover the power needs of an A-Class vs. a comparable Series 9x9 HP 3000. We were prompted by a roadshow talk that HP delivered today here in Austin. Part of the content touted HP's cooler hardware designs, and we don't mean "more hip" when we say cooler.
We mean power and cooling energy efficiency, a measurement that ranges from the wattage a CPU draws to the needs of a blade server or blade storage, up to the electricity required to keep a full server enclosure running. Austin was a good place to have the talk, since we've posted 82 days of 100 degrees or more this summer, blasting our own record from 1925.
HP's solutions "span across IT and facilities to optimize and adapt energy use, reclaim capacity, and to reduce energy costs." Sullivan's Steakhouse lunchtime diners heard about the latest advances in power protection, distribution, and cooling. HP showed the numbers on calculating operational costs "to help extend the life of the datacenter."
Datacenters are migrating in the 3000 market. We're polling the community about their career and company changes over the 10 years since HP pulled out of the market. Some switched off 3000s because of high power needs. A new case history by MB Foster about decommissioning a Series 969 3000 at San Mateo Health Care cites cutting power as a spark to get off a 3000.
Comparing power needs requires great access to hardware manuals, which are genuinely useful in PDF format. Before it pulled away from your market, HP always crowed about slashing the power needs of a 3000 with the PCI-based systems introduced in 2001. The power-efficiency of Integrity blade systems running in, for example, a C3000 enclosure (the smallest) is even more pronounced over those 9x9 servers.Brian Edminster has been our go-to fellow for a host of news stories recently, including his own Applied Technologies open source repository for software that can help HP 3000 owners. He says that power and heat specs for 9x9 vs. A- or N-Class 3000s can be found in an HP datasheet "that spells out 'basic' specs for A-Class systems and the same for N-Class systems. A site prep guide for 9x9 servers is online at ManualShark, at http://www.manualshark.org/manualshark/files/28/pdf_26712.pdf
Edminster warned us that "the power use ratios are very rough-cut based on 'system chassis' only (simply CPU, memory, and internal disk/tapes). They are basically the smallest system that can boot MPEi/X.
The data I've provided the weblinks to is only a rough guideline, but enough to show what HP was talking about regarding the power consumption: significantly lower with PCI-based systems. The A and N-class guides are just for the system proper (no external expansions). You can dig the same data out of that 9x9 manual, but it's harder to get at.
In short, HP was right, at least about power requirements. The ratio of power usage (based on A-Class being very roughly = 1) is A=1, N=3, 989=6. You really don't even want to know about the 99x systems. Another point to consider is that peripherals (especially external disk arrays) can really be power hogs too -- and the newer arrays (with fewer but larger drive mechs) had similar power usage ratios.
A fully built-out multi-rack Series 989 system would likely have in excess of a 10:1 power draw compared to a roughly equivalent N-Class, which would only take about half a rack of space. So if space, power, or HVAC capacity are at issue in your datacenter, newer is better.
Edminster added that he's curious how the new Stromasys HPA/3000 CHARON systems will fall in, "given that they'll be based on pretty much cutting edge hardware." The emulator was running on a gaming PC with SSD storage at the HP3000 Reunion in September. You couldn't even hear a fan.
November 08, 2011
A decade later, Windows XP still in business
Last week an Internet tracking company reported that Windows XP, first released in 2001, was losing market share at a record pace. The decade-old software is now running on just under half of the world's market share of Windows systems.
The 2.5 percentage points which XP lost in October leaves the OS with a 48 percent share. It was running 55 percent of the Windows PCs as recently as August, according to Net Applications.
Microsoft would like to insist that its customers stop using an OS that was last enhanced in 2004, with the arrival of Service Pack 3. But SP3 is what's keeping companies in the Microsoft fold, just like MPE/iX enhanced in 2007 continues to provide HP with some targets for limited support services. HP can't force a company to stop using an OS last updated, well, more recently than Windows XP. A solution that continues to get the job done usually remains in place, until an acquisition goes down, or something breaks.
What's most likely to break in Windows XP SP3? There's one element that's certain to stop running: Microsoft's patching service for the OS, in 2014. Yes, the formula performed by HP upon the 3000 is a plan familiar to the rest of the IT industry. Time marches on for software which the vendor considers outdated. Internet Explorer, not exactly the most robust calling card, will not include an IE9 for XP's users. Of course, there's other browsers available to the Windows user, such as Google's Chrome, or Firefox. IE is required to run Windows' interface, however.Browsers are a big issue in charting the lifespan for the version of Windows more widely installed than any other. (Vista holds a 9 percent share, while Windows 7 gets the rest.) In an article in Computerworld, the XP situation is painted based on colors from a survey by Net Applications, which uses data "obtained from more than 160 million unique visitors who browse 40,000 websites that the company monitors for clients."
Another group that operates for clients is the Gartner Group, whose Michael Silver reported that enterprises don't want to run an operating system when there's no security fixes available. The Windows XP environment is very unique in the computer industry, however. It's a proprietary system -- nobody's got source code for Windows, except Microsoft -- that drives, well, half of the greatest share of the world's desktops and enterprise servers.
In contrast, HP licensed the source code for MPE/iX to seven companies this year. Most of them haven't needed to use it yet, but we've heard that the source code has been quite valuable to the makers of the emulator that could be slowing down migrations in the 3000 space.
There's a migration in the future for companies which are adopting a Windows alternative to using HP 3000s. At least any of those who will find themselves the 48 percent of the XP-using populace. Microsoft wants to migrate everybody to Windows 7, now that the Vista experiment hit the wall with the user base. Windows 8 will have a smaller shot at the migrating market; Silver said the Windows world has "migration fatigue."
Windows XP users have almost three more years to get their migrations finished before those Microsoft patches disappear. There's been requests from the customers to get XP extended a year or two -- something like what MPE/iX and the 3000 got, as a lease on support life from HP in 2006. And in 2008. Microsoft is likely to say no, according to Gartner's Silver. But about 10 percent of that 48 percent don't see any urgency in migration, or haven't done anything to prepare, he added. Microsoft doesn't want any part of keeping those customers on XP.
"The longer they let them run XP, the more enterprises will slow down their migration," Silver said in the article.
The pace of migrations is a serious issue to the 3000 community suppliers who offer alternatives to the environment. A rise of emulator news and promises is likely to "slow things down for awhile," according to ScreenJet's Alan Yeo.
But the vendor's ultimatum date of Dec. 31, 2010 has already passed. So migrations from this point onward in the 3000 community are being spurred by something more compelling than support via patches -- which most system managers avoided anyway, according to support companies such as Pivital Solutions, or Beechglen.
What is triggering the remaining migrations is application aging, as well as the aging of the experts who can maintain and update them. When it becomes easier to hire and train Windows experts on the business logic of HP 3000 sites than to find MPE managers, a company has a strategy to make a switch. This is where HP is being cut out of the picture during migrations -- the spot where another vendor's proprietary operating system, not HP-UX, takes over for the 3000.
10 years of service in the PC community is a lifetime, according to the Gartner experts. And yet XP lives on, driving back the pace of change until this summer. Given enough time, and the continuity of expertise in Windows, a genuine majority of Windows 7 users is not far away. The continuity of experience helps keep the Windows customers in the fold, even through a migration. HP has not been as fortunate, telling the chief of the Stromasys emulator technical group it lost half of its 3000 sites to other vendors in a six-year period of migrations.
November 07, 2011
Emulator posts demonstration website
Stromasys announced that it won't take on any more North American test sites for the CHARON-HPA/3000 emulator. The company building the first virtualized hardware server that acts like PA-RISC systems wants to provide the best test experience for companies already in the program.
To properly support our HP 3000 Field Testers, we have stopped accepting participants in North America. We can still accept several test sites in Europe. We very much appreciate your support, and we received excellent feedback on problems. Most problems relate to configuration and installation procedures, so we are updating our documentation.
Starting Nov.8, the 0.3 version of the software will be available to download for test sites from the Stromasys servers. But most important to the rest of the community, a demonstration system is accessible via on the Stromasys website. "The system is available 24x7 and intended for general information," according to the latest field test newsletter.The company has added support for virtual tapes larger than 4GB in the 0.3 release. There are also fixes to the real-time clock, which deviated a few minutes per day. The engineering staff also worked out an issue with heavy loads on SCSI interfaces to the software-hardware combo. Field testers now have an example tape image to download, but "most testers have now produced their own [disk] images."
There's a new full-time engineer working in the development group, which is led by Igor Abramov out of the Moscow office. The extra help will "accelerate the development of a fast PA-RISC virtual CPU model and on SMP system configurations," the company said. A ShenZhen test lab runs CHARON-HPA/3000 testing in parallel, preparing for product support in Asia-Pacific.
"We are on target for a first product release in January, 2012," the field test note adds.
November 04, 2011
Several seats now open on OpenMPE board
It's been about six weeks since we had anything to report about OpenMPE. We've promised not to spread any more posts unless there's news. Several directors have resigned their positions in the volunteer group. Birket Foster has returned to the chairman's spot, due to one of the resignations.
Foster said that Jack Connor, who took on the chairman responsibilities early this year, decided he'd had enough of his volunteering for the homestead community. He joined the group in 2010 and led initiatives including a salable product and a spring fundraising effort. Connor's is one of three departures from the group this fall, starting with vice-chairman Keith Wadsworth in September. (Wadsworth came on board with Connor, as well Connie Sellitto.) Tracy Johnson is also not listed among the board directors on the group's website today.
Connor, who's been the most public to responding to queries about OpenMPE, updated the finances of an entity that wanted to raise $50,000 from the community this spring. "A great deal of the current funding has come from loans and contributions from the board members," said in March, when the fundraising was fresh. "This shows their commitment to the OpenMPE concept."
That concept has changed constantly over the almost nine years of its existence. It began with the desire to get HP to open up source code and technology about a server it was discontinuing. Then the mission was the creation of a 3000 emulator. MPE/iX licensing issues, as well as necessary but overlooked HP procedures for migration, because the longest-term mission.
This fall, the group is maintaining a 3000 server at Measurement Specialties, with a DR backup at The Support Group, a pair of donated Series 9x9 3000s that host useful software for homesteaders. 2011 was the Year of the Lawsuit for this group, legal action sparked by former treasurer Matt Perdue after he was voted out over a failure to pay a bill. He named Connor and Wadsworth in his lawsuit, along with the board as a whole. The legal expenses to respond in a Texas court were significant enough to consume much of the group's energy. A settlement was arranged, but its details have been sealed. Throughout 2011, monies listed as "contributions" were coming from these volunteers, in the majority. Some were considered loans, and some of those have not been repaid to the board members.
However, OpenMPE has more than board donations and resignations, an attempt to license its own source code copy of MPE/iX for development, and legal action as its legacy to date. This is also the group that pressed for the release of that MPE source code during 2002-08. Much more significantly, these volunteers made HP see the need for an emulator license for MPE/iX in 2005. That license is now essential to making the Stromasys HPA/3000 emulator a product ready for sale in a few months.Foster updated us on this fall's state of this volunteer group, including finances. "The only thing we have to repay is legal bills," he said. "Everything else has been covered by the small donations that have come in from individuals." (Full disclosure: the NewsWire was among these contributors, sending $500 in March 2010 -- the month when the HP bill came due for source licensing.) Foster also mentioned Gilles Schipper of support provider GSA, plus we know that Brian Edminster of Applied Technologies contibuted $1,000.
Combined with the Measurement Specialities and Support Group contributions of maintenance, administration and rackspace/power, plus servers from Client Systems, Foster says this group has retained the advocates for its missions. "It's helped, because people do believe this organization needs to exist for advocacy," Foster said. "Our most recent objective was to get the source code released, and that worked," he added. Seven companies now have source code licenses for the portions of MPE/iX that HP was able to license.
The history of OpenMPE has shown a lot of caution and detailed debate right from its beginnings, two months after HP curtailed the vendor's 3000 futures in 2001. Looking at the online minutes, there were concerns over using "Open" as well as "MPE" from board members at the very first meeting in January, 2002. Despite the confidentiality demand HP insisted upon, "we also did the advocacy for the transfer of licenses, and to include an emulator in that license transfer" through 2008, Foster notes.
The critics of this group -- and it has had them ever since HP made OpenMPE go into nondisclosure in 2004 -- say that nothing has emerged from OpenMPE but talk and promises. But if you believe HP wanted to create an emulator license on its own, or think some indie vendor made that happen with no OpenMPE help, you may as well believe believe HP's TouchPad is going to be a big hit this Christmas.
In some isolated places among the community, there have been calls to shut OpenMPE down. These are requests for closure of an office that doesn't exist, director compensation that never emerged, as well as a treasury that couldn't report a five-figure balance at any time over the last eight years. The assets of OpenMPE today appear to be volunteers' work on those two donated servers; the liabilities, legal fees unpaid.
Within two months of Wadworth's arrival on the board, he initiated his own probe into the group's by-laws and corporate organization. "His first report back to us was that we were improperly constituted," Foster says. "In the long run it turned out we were not improperly constituted," because the state of Maryland where the group was launched permits a non-shareholding, for-profit company. Wadsworth also introduced suggestions that led to creating Corporate Structure, Mission Statement, Membership Definition and Communications sub-committees. In the spring 2010, this group had nine members on its board. By the end of summer it could count seven, after Sellitto's departure as well as Ann Howard's.
After a fall 2010 that looked busy with setups of the community servers, and a report of a $5,600 balance, the trail of activity of OpenMPE drifts into executive sessions, starting with the dismissal of Perdue. A December note reports that Texas Bexar "County declined proscecution of [Perdue,] as the amount was too small a figure." Wadsworth's Orbit Software had paid legal fees to respond to the suit, so the other board members agreed to "contribute $1,000 each to repay Orbit due to fiduciary loss."
Within two weeks of the last December, 2010 meeting, openmpe.org went offline for several weeks. The board consisted of six members at the start of 2011. A fresh bank account had been opened and the group received a new invoice from Orbit. The January 6 minutes report a "Discussion of future purpose and organization of OpenMPE," and authorized consultants to be contracted.
Later that month, the group drafted a letter notifying HP that it had lost its copy of the source code, which was sent to Purdue's offices in 2010. Connor was voted in as new chair, a ballot cast among the other five board members. By February a call for fundraising emerged at the same time that the annual election of directors was deferred "until such time as we have completed changes to our by-laws, which will allow us to define membership, goals and mitigate risks for board members."
After a final $1,000 reported contribution, this April began with the group "soliciting opinions from former members of the board regarding a business plan. Three items to be included as salable products: INVENT3K subscriptions; an HPSUSAN change tool [NoWait/iX, created by Connor]; and commisioned sales of 3000 upgrades via Client Systems, which has donated servers to the group."
None of this group's critics have made a case for how the community could benefit from a shutdown. Invent3K could count 18 subscribers in April, of which just two had paid a $99 fee. Meeting minutes for April and May don't contain any details of group actions, although executive sessions are noted during this period when Perdue's lawsuit was settled. OpenMPE's reports of minutes go dark after May 12 on its website. No volunteers have joined as directors since March of 2010. Tony Tibbenham and Alan Tibbetts remain listed on the website as directors along with Foster.
November 03, 2011
Is it a test next Wednesday, or the real thing?
By Birket Foster
MB Foster Associates
Remember growing up where a test of the emergency alert system would periodically come on the radio with a piercing noise, and inform you that this was just a test, and in the case of a real emergency further instructions will follow? It's coming back, sooner than you think. We expect to hear it just minutes before our next webinar.
For the first time ever, the FCC and FEMA are conducting a nationwide test of the Emergency Alert System (EAS). It will happen at 2 pm Eastern on Wednesday, November 9 -- just 2 days before binary day (you know, 111111, or November 11, '11).
Will our children even know what this is about, as they get home from school and have their favorite TV show interrupted by an emergency signal? Will housewives everywhere worry about the “national emergency” that will appear to be happening? Maybe it will have the same impact on society as the radio play of HG Wells “The War of the Worlds,” a timeless science fiction classic of the invasion of earth by Martians.
Our live demonstration of MBF Scheduler, and the recently announced enhancements of HIPRI, RUNNOW and Subqueues, will take 45 minutes -- but during the first two minutes we will see how the world reacts to EAS, and then carry on from there.Wikipedia reports that the EAS is a national warning system in the United States put into place on January 1, 1997, when it superseded the Emergency Broadcast System (EBS), which itself had superseded the CONELRAD System.
In addition to alerting the public of local weather emergencies such as tornadoes and flash floods, the official EAS is designed to enable the President of the United States to speak to the United States within 10 minutes, but the nationwide federal EAS has never been activated. The EAS regulations and standards are governed by the Public Safety and Homeland Security Bureau of the FCC. Each state and several territories have their own EAS plan. EAS has become part of IPAWS, the Integrated Public Alert and Warning System, a program of Federal Emergency Management Agency (FEMA). EAS is jointly coordinated by FEMA, the FCC, and the National Weather Service.
So next week for two minutes the airwaves of America (which are broadcast by satellite around the planet) there will be a two-minute alert where the president or at least the White House will interrupt the day for 2 minutes to prove that in the case of the a national emergency you could hear the President – it would just be text and a voice over – no pictures … maybe the lowest common denominator isn’t the best way to go but all radio and TV will be part of the test.
How will you business handle this – are you forewarning people that this test is coming up … At MBFoster we have a webinar scheduled for November 9 at 2pm … it will be about Automating Windows Processes (using MBF Scheduler) … it won’t be a test – it will be the real thing J complete with demonstration … it will take 45 minutes but during the first 2 minutes we will see what the world reaction to the of the EAS is and then carry on from there.
November 02, 2011
For big power savings, HP thinks smaller
HP announced yesterday that it is testing chips from an Austin, Texas mobile processor company for use in Hewlett-Packard servers. Calexda is one of a host of chipmakers who produce ARM processors. HP said it believes these chips can provide the needed horsepower for server tasks -- while sipping power, instead of gulping it like anything from Itanium to Xeon chips.
HP is very serious about reducing power consumption at its enterprise customer sites. The vendor has a road show in play that addresses this benefit of moving off older HP 3000 hardware. The PCI-based N-Class and A-Class servers reduced power consumption (as measured by BTUs) by 30 percent over the 9x9 Series. And the Integrity 2660 class of servers, similar to an A-Class, use 567 watts at idle to support an entire server.
A mobile chip solution for enterprise establishes a fresh measuring tape for power usage. HP is calling the initiative to create a new server line Project Moonshot. It hopes to start selling these ARM-based servers by next year. The rollout at the Calexda HQ in Austin showed off the EnergyCore ARM system-on-chip (SoC) for cloud servers and on-demand processing.
Hewlett-Packard isn't planning to introduce hardware for its small to midsize customers anytime soon that utilizes EnergyCore ARM. The math on the pricing will not help HP's revenue numbers, if it was deployed all the way down the customer lineup. A load that normally requires a $3.3 million system of 400 servers, with 10 storage racks and 1,600 networking and power cables using 91 kilowatts of power, could be done in the new system for $1.2 million: using one-half a storage rack, 41 cables and 9.9 kilowatts. Those are enticing number for cloud compute suppliers like the emerging manufacturing alternative Force.com. HP said these kinds of customers will be looking at system-on-chip solutions -- which could drive down the costs of cloud computing for the masses who might be migrating.HP's Redstone Development Platform is the hardware end of the Project Moonshot, something HP describes as "the industry’s first server development platform to feature extreme low-energy server processors that consume almost 90 percent less energy."
HP is tying Redstone into its HP Converged Infrastructure to help cloud providers share technology resources across thousands of servers. The technology targets the future of "low-energy computing for emerging web, cloud and massive scale environments."
In yesterday's podcast about the "post-HP" era, we pointed out how much spadework and mucking out the company must do to become relevant once more as an enterprise alternative to the 3000. Reducing power needs by 90 percent -- for anyone on the IT power food-chain -- qualifies as a leap out of the mud of "do we keep selling PCs, or not?"
If IT planning involves the selection of a vendor for your migration -- or simply a check to see if your application can operate on HP's environments -- this ground-breaker might deserve some time for a closer look, just to have something to quiz your cloud supplier about. One place to start is the "media kit" for its Low Energy Server Technology. What drives these innovations is a long way upstream from the architecture of a local server running MPE/iX applications. But this is the future where HP is investing in its hardware, "warehouse computing." An IEEE paper explains that power is one of the most critical concerns in this new concept.
One of the biggest trends in the server market has been the emergence of the large-scale data center, driven by Internet-sector growth. Indeed, recent market research identifies the Internet sector as the fastest-growing segment of the overall server market, growing by 40 to 65 percent every year, and accounting for more than 65 percent of low-end-server rev- enue growth in 2006. Furthermore, several recent news articles and keynote speeches have highlighted this area’s importance.
One of the most interesting aspects of this growth is the unique set of challenges it presents for server design. Internet-sector infrastructures have millions of users running on hundreds of thousands of servers, making the ability to scale-out server configurations a key design requirement. Experts have compared these environments to a large warehouse-style computer, with distributed applications such as mail, search, and so on. For companies in this market, data-center infrastructure — including power and cooling — can be the largest capital and operating expense, motivating companies to focus on the sweet spot of commodity pricing and energy efficiency.
Another IEEE paper points out the problem that the ARM-based Moonshot must overcome: latency. Software has to be explicitly parallelized to run effectively across so many "wimpy" processors, which adds to development costs. HP's going to open up a Discovery Lab in Houston in January and invite developers to test its new technology.
But they'll need to be clever to keep the feet of the IT horsepower moving on lower power.
The IEEE paper cautions that the overhead of splitting things up across multiple processors can reduce performance; and there’s the likelihood of lower utilization of the processors, and thus a loss of efficiency.
Once a chip’s single-core performance lags by more than a factor to two or so behind the higher end of current-generation commodity processors, making a business case for switching to the wimpy system becomes increasingly difficult because application programmers will see it as a significant performance regression: their single-threaded request handlers are no longer fast enough to meet latency targets. So go forth and multiply your cores, but do it in moderation, or the sea of wimpy cores will stick to your programmers’ boots like clay.
November 01, 2011
Listen for the sounds of a post-HP season
In less than 10 minutes of our latest podcast, we're connecting the dots on Steve Jobs, his reverence for HP, the company's PC reverse-march, and how much Hewlett-Packard lost while it exited the 3000 market. It all points to a chilly off-season while HP works to get back onto the field of enterprise computing, carrying its PCs, and take another run -- like the Texas Rangers -- at the Number 1 spot.
Post-HP? For awhile, anyway. On this first day of its 2012 Fiscal Year, HP is working away from a year when it couldn't seem to get a strike when it needed it, either off the bat of CEO Leo or from the arms of its TouchPad. Maybe it's time that we stop looking back at what HP didn't do a decade ago -- like stick to a profitable, small HP 3000 business. Or stay out of a slim-margin dogpile like the PC business. Or remain focused on enterprise computing. As they say in baseball -- especially here in Texas -- there's always next year.