Adventures in systems land

Yep, it’s a mouthful. I’ve just been booking some events and presentations for later in the year, and this one, which I had initially hoped to attend clashes with one, so now I can’t.

However, in case the snappy new title passed you buy, it is still the excellent IBM Technical conference it used to be when it was the IBM System p, AIX and Linux Technical University. It runs 4.5 days from 8 - 12 September in Chicago and offers an agenda that includes more than 150 knowledge-packed sessions and hands-on training delivered by top IBM developers and Power Systems experts.

Since the “IBM i” conference is running alongside, you can choose to attend sessions in either event. Sadly I couldn’t find a link for the conference abstracts, but there is more detail online here.

I got an email pointing out that I omitted a link to the youtube video of the IBM hydro-cluster. So, here it is.

Towards the end of the video, Jeff Gluck says “hot water can be moved off site”, “to heat your home or cook a family dinner”. In the famed Larry and Brin, “do no evil” context, I guess this is goodness. While I appreciate that there is a very serious side to the “greening” of the datacenter, I couldn’t help but laugh.

Back in the 1970’s on one of the first large scale computer servers, aka mainframes I worked on, we used to store takeaways inside the server for 4-5 hours to keep it warm on evening and night shift. The really scary thing, back in those days microwaves didn’t exist!

The IBM 370/145 was a T-shaped server, laying on its back, the whole back of the T was largely empty, ready in case you wanted to upgrade to a 370/148 or 155(I think). So it became common place to store stuff in there that you wanted to keep warm and dry. Ideal for takeaway and girlie magazines(so I’m told!).

I’ve had a few emails asking me if I was going to write a log entry on this month announcements, and to be honest I wasn’t. They are an impressive list of products, branding and customer announcements. I wasn’t anything to do with them, given I’m no longer asked to do marketing/sales types presentations, I picked that time to go do the Machupichhu/Inca trail trip in Peru.

The April announcements though were a credit to the teamwork across the even more global IBM. Core Processor and server development teams in Austin and Rochester, worked with domain specialists in Poughkeepsie and Boeblingen. On top of this were the software development and test teams in India, China and and ever increasing number of places.

The new UNIX enterprise server, the Power™ 595 is an impressive beast if the charts are anything to go by. I’m hoping to get Nancy to take me across the building to the test bring-up to have an up close and personal look sometime this week. The new POWER6 “Hydro-Cluster” supercomputer, the Power 575, is very impessive using a new super-dense system, with a unique, in-rack, water-cooling system and with 448 processor cores per rack. Apprently it offers users nearly five times the performance and more than three times the energy efficiency of its predecessor, IBM’s POWER5+™ processor running upto a industry busting clock cycle of up to 5 GHz.

These Super-dense systems are starting to become a really interesting value prop. On Friday I got a link to the IBM.COM public website that included a video on our iDATAPLEX offering. It was there Saturday and has gone today, but it was there as this search in the current google index shows. The video doesn’t show any technical details but does give an interesting insight into this x86 based super-dense, Internet scale, behemoth of a server. I was hoping there was other public comment or blog entries I could leach off for discussion points, but the only search results return job postings

Anyone go to the iDATAPLEX session at IMPACT 2008 and want to comment ??

That got your attention didn’t it?

We’ve announced another performance and benchmark record this week, IBM WebSphere Application Server benchmark involved more than 109,850 concurrent clients and produced 14,004.42 SPECjAppServer2004 JOPS@Standard (jAppServer Operations Per Second), which translates into more than 50 million business transactions over the course of the benchmark’s hour-long runtime. That’s a lot of clients, and a lot of transactions!

The performance run was completed on IBM POWER6 BladeCenter servers powered by two dual-core IBM® POWER6® 4.0 GHz processors and IBM DB2 Universal Database v9.5 on a System p p595 running AIX.

We ran the test over 52-processors, 2-cores per processor and with SMT on. The software config included 26 WAS instances. Now, the issue here isn’t performance, 26-instances isn’t so bad from a config and deployment perspective either. But wouldn’t it be better if you could bundle that all up into a couple of racks and use cloning, automatic deployment, recovery, scheduling etc. and on an even more consolidated, energy efficient platform.

Funnily enough, we are working on that. The IBM Press release mentions IMPACT 2008, that might be good timing, I won’t be there as I’m off to do the Machu Picchu thing at the start of April.

Prior to the new WebSphere+Power double-up, the 4Q2007 record was held by Oracle on HP-UX Integrity Server Blade Cluster, with 10,519.43 JOPS over 24 server instances on 22 2-core processors; Sun also submitted a SPARC T5120 SPECjAppServer2004 benchmark with Sun Java System Application Server 9.1 running 6-nodes, 18-server instances on 48-cores, 6-chips and only scored 8,439.36 JOPS.

You can read the full press release with links to SPEC and IMPACT 2008 here.

Feb. 12th Datamation announced their product of the year awards, the IBM Power Systems p570 server won enterprise server of the year, up against the IBM System x 3950 M2 Server, the HP MediaSmart Server, and the Dell PowerEdge 2970.

Details on all the award winners are here.

Liz Stahl has a good summary on the performance, benchmarks etc. on this weeks IBM System p 520 and IBM System p 550 server announcements.

Yesterday saw the announcement of a re-packaging, re-branding and new technology drive for POWER™ Virtualization now PowerVM™. You can see the full announcement here. It is good to be back working on VM, sorta.

Over on virtualiztion.info, Alessandro Perilli, says we are “missing the market in any case because its platform is unable to virtualize Windows operating systems”. I say not.

POWER isn’t Windows, it’s not x86 hardware. We scale much, much higher, perform much better and generally offer high availability features and function as standard or an add-on, way ahead of Windows. Running Windows on PowerVM and Power hardware would pick-up some of the reliability features of the hardware transparently, and the workload consolodation potential would be very attractive. What it comes down to though, is what it would take to virtualize Windows on PowerVM?

We could do it. We could add either hardware simulation or emulation or more likely translation that would allow the x86 architecture or Windows itself to be supported on PowerVM. There would be ongoing issues with the wide variety of h/w drivers and related issues, but lets put those aside for now.

We could have gone down a similar route to the old Bristol Technologies WIND/U WIN32 licensing and technology route, porting and running a subset of WIN32 or even via mono or .net. We might even call it PowerVM Wx86. Just reverse engineering MS technology is neither the right idea from a technology or business perspective.

So technically it could be done one way or another. The real question though is the same as the discussion about supporting Solaris on Power. Yes, it would be great to have the mother of all binary or source compatibility virtualization platforms. However, as always the real issue is not if it could be done, but how would you support your applications? After all isn’t it about “applications, applications, applications“?

And there’s the rub. If you wanted to run middleware and x86 binary applications on POWER hardware, then you’d need support for the binaries. For middleware, most of the industries leading middleware is already available on either of AIX, i5/OS or Linux on Power, some is available on all three. What would software vendors prefer to do in this case? Would they be asked to support an existing binary stack on Windows on PowerVM, or would they prefer to just continue to support the native middleware stacks that benefit directly from the Power features?

Most would rather go with the native software and not incur the complexities and additional support costs of running in an emulated or simulated environment. The same is true of most customer applications, especially those for which the customer doesn’t have easy or ready access to the source code for Windows applications.

In the x86 market, the same isn’t true, there’s less risk supporting virtualization such as Xen or VMware

The same isn’t true with PowerVM Lx86 applications. First because of the close affinity between Linux and Linux on POWER. There are already existing Linux on Power distributions, the source code is available, and most system calls are transparent and can be easily mapped into Linux on POWER. Second, drivers, device support etc. is handled natively within either the POWER hardware, PowerVM or within the Linux operating system, running in the PowerVM partition. Thirdly, IBM has worked with SuSe and RedHat to make the x86 runtime libraries available on Linux on POWER. Finally, many middleware packages already run on Linux on POWER, or it is available as open source and can be compiled to run on Linux on POWER.

All of which makes it a very different value proposition. Using NAS or SAN storage, it is perfectly possible to run the same binaries currently or as needed on x86 and PowerVM. The compilcations of doing this, the software stack required, as well as the legal conditions for running Windows binaries just don’t make it worth the effort.

Although not identical, many of the same issues arise running Solaris, either Solaris x86, or OpenSolaris PowerPC port. So, thats a wrap for now, still many interesting things going on here in Austin, I really well get back to the topic of Amazon, EC2 and loud computing, memo to self.

Trying to find a reference book on AIX 6, I looked at the latest list of Redbooks for Power Systems, these are the ones listed in the RSS feed since the start of October 2007.

Continue reading ‘Catching up on IBM Redbooks’

Simulation is a big deal these days, not only can you get well ahead of actual chip availability by simulating architecture, but you can also run where otherwise you couldn’t normally.

Last week IBM Global Engineering Services (GES) inked a deal with Virtutech to use their Simics™ product to create advanced, system-level simulation models for IBM’s next-generation embedded processor cores, beginning with the PowerPC 464FP core.

More details here in the Virtutech press release.

Thanks to the folks over at the “Power Architecture zone editors’ notebook” blog here is their summary of last weeks announcements.

Get yours today: Listen UNIX users — the newly available IBM BladeCenter JS22 with Power6 is what you’ve been waiting for. Couple the JS22’s Power6 processor technology with the built-in Advanced Power Virtualization and you’ve got a lot of Power concentrated in a compact container (which can also save you on space and energy costs). It comes loaded with two 4GHz dual-core processors, an optional hard drive, and an Ethernet controller; it supports as much as 32GB of memory; the first shipments are configured for the BladeCenter H and BladeCenter HT chassis. And its virtualization features make it really special (see following entry for more on this).

And what’s a new blade without a complementary OS: Targeted for Friday, November 9, 2007, the release of AIX 6 from the beta bin should provide users improved security management and virtualization features that take advantage of a hypervisor included in the Power6 processor so you can get 100 percent application up time. The Workload Partition Manager should let sysadmins create multiple partitions (each with customized memory settings for users and application workloads) and the Live Application Mobility feature can shift applications from one server to another on the fly (and they keep running while migrating). Then there’s the Security Expert which lets users control more than 300 security settings (role-based access to applications, user-based authentication, etc.). These OS utilities should work well with the Power6 Live Partition Mobility hypervisor which can move an entire OS (AIX, RHEL, and SLES) and its workloads from one server to another while they are running. (In fact, you can preview AIX 6 here if you can’t wait until Friday.)