zEnterprise Workload Economics

IBM never claims that every workload is suitable for the zEnterprise. However, with the advent of hybrid computing, the low cost z114, and now the expected low cost version of the zEC12 later this year you could make a case for any workload that benefits from the reliability, security, and efficiency of the z is fair game.

John Shedletsky, VP, IBM Competitive Project Office, did not try to make that case. To the contrary, earlier this week he presented the business case for five workloads that are optimum economically and technically on the zEnterprise.  They are:  transaction processing, critical data workloads, batch processing, co-located business analytics, and consolidation-on-one-platform. None of these should be a surprise; possibly with the exception of analytics and consolidated platform they represent traditional mainframe workloads. DancingDinosaur covered Shedletsky’s z cost/workload analysis last year here.

This comes at a time when IBM has started making a lot of noise about new and different workloads on the zEnterprise. Doug Balog, head of IBM System z mainframe group, for example, was quoted widely in the press earlier this month talking about bringing mobile computing workloads to the z. Says Balog in Midsize Insider: “I see there’s a trend in the market we haven’t directly connected to z yet, and that’s this mobile-social platform.”

Actually, this isn’t even all that new either. DancingDinosaur was writing about organizations using SOA to connect CICS apps running on the z to users with mobile devices a few years ago here.

What Shedletsky really demonstrated this week was the cost-efficiency of the zEC12.  In one example he compared a single workload, app production/dev/test running on a 16x, 32-way HP Superdome and an 8x, 48-way Superdome with a zEC12 41-way. The zEC12 delivered the best price/performance by far, $111 million (5yr TCA) for the zEC12 vs. $176 million (5yr TCA) for the two Superdomes.

When running Linux on z workloads with the zEC12 compared to 3 Oracle database workloads (Oracle Enterprise Edition, Oracle RAC, 4 server nodes per cluster) supporting 18K transactions/sec.  running on 12 HP DL580 servers (192 cores) the HP system priced out at $13.2 million (3yr TCA). That compared to a zEC12 running 3 Oracle RAC clusters (4 nodes per cluster, each as a Linux guest) with 27 IFLs, which priced out at $5.7 million (3yr TCA). The zEC12 came in at less than half the cost.

With analytics such a hot topic these days Shedletsky also presented a comparison of the zEnterprise Analytics System 9700 (zEC12, DB2 v10, z/OS, 1 general processor, 1 zIIP) and an IDAA with a current Teradata machine. The result: the Teradata cost $330K/queries per hour compared to $10K/queries per hour.  Workload time for the Teradata was 1,591 seconds for 9.05 queries per hour. That compared to 60.98 seconds and 236 queries per hour on the zEC12. The Teradata total cost was $2.9 million versus $2.3 million for the zEC12.

None of these are what you would consider new workloads, and Shedletsky has yet to apply his cost analysis to mobile or social business workloads. However, the results shouldn’t be much different. Mobile applications, particularly mobile banking and other mobile transaction-oriented applications, will play right into the zEC12 strengths, especially when they are accessing CICS on the back end.

While transaction processing, critical data workloads, batch processing, co-located business analytics, and consolidation-on-one-platform remain the sweet spot for the zEC12, Balog can continue to make his case for mobile and social business on the z. Maybe in the next set of Shedletsky comparative analyses we’ll see some of those workloads come up.

For social business the use cases aren’t quite clear yet. One use case that is emerging, however, is social business big data analytics. Now you can apply the zEC12 to the analytics processing part at least and the efficiencies should be similar.

New Announcements Fuel zEnterprise Rebound

A couple of weeks ago DancingDinosaur looked at the year-end results posted by IBM, with the z, led by the zEC12, leading the gains. Despite some tough quarters earlier in 2012, the System z has been on quite a roll; in 4Q12  it experienced 56% year to year revenue growth, the strongest since 2000.

The numbers look that much when better compared to what had immediately preceded them. In 3Q12, the z was down 19% on revenue (ouch).

The financials prove what long-time IBM watchers know: a new mainframe intro always kicks up the numbers, although it may take a quarter or two. The new machine, the zEC12, launched at the end of August 2012. Historically IBM follows a new mainframe with a business class version about a year after the initial launch, so we can expect a business class version of the zEC12 around August or September. (The z114 was introduced in July 2011, a year after the z196.) You can expect IBM will price it aggressively, as was the z114.

But it wasn’t only the new machine driving the good number.  Since the introduction of hybrid computing in 2010 with the zEnterprise 196, IBM has been doggedly pushing the idea of cloud computing, analytics (big data or otherwise), efficient centralized management, elastic scalability, and mobility. It has steadily brought out tools and optimizations that enhance the z for these kinds of workloads. Today its three main growth initiatives for the z focus on cloud, analytics, and security, which parallel IBM’s overall Smarter Computing thrust.

Optimizations and enhancements, like those announced or previewed earlier this week, will continue to drive the zEnterprise forward. For example, newly enhanced tools like z/OSMF (z/OS Management Facility) ease management of z/OS through a modern, browser-based console for z/OS.

The newest version, z/OSMF Version 2.1, previewed this week, provides intuitive management intended to enable IBM’s Smarter Computing through such capabilities as at-a-glance reports on software service levels and system software assets. New workflow capabilities promise to simplify z/OS configuration tasks and tune them to user roles. It leverages the Liberty profile for WebSphere for z/OS to accelerate deployment and speed time to value. Finally, it includes a REST API for jobs submission, which bridges batch and web-based applications and, in the process, helps bring younger z/OS system programmers up to speed.

Eletrobras Electronuclear, Brazil, the largest generator of electricity in Latin America, turned to z/OSMF for help in cloning its production system to ensure system resilience. As an Eletrobras manager noted: With z/OSMF this cloning process is much easier and can be completed in a matter of a few hours.

Also previewed was z/OS Version 2.1, described by IBM as the Smart Foundation for Smart Computing. z/OS v2.1 brings enhanced security through a new crypto server and support for additional industry security standards.  Enhancements around paging and throughput enable greater scalability while providing optimized performance reporting.  It offers enhanced data serving to fuel analytics for new data-hungry analytic applications as well as optimized data storage and faster file retrieval through an enhanced zFS along with improved FICON management. For management, it includes built in workload optimization, extended management reporting, and continued batch modernization.

IBM also previewed a z/VM upgrade, version 6.3. This adds support for 1TB of real memory and improved performance with HiperDispatch. The z/VM memory will provide better performance for larger virtual machines and reduced LPAR sprawl by allowing up to 4x more VMs per LPAR.  z/VM v6.3 boasts a higher server consolidation ratio with support for more virtual servers than any other platform in a single footprint, according to IBM.

Meanwhile IBM continues to roll out z customer examples at an increasing pace. Bankia, a Spanish banking company, turned to the System z for real-time integration of data to facilitate monitoring and mitigate risk. Vantiv, a leading payment processor based in Cincinnati, turned to the z for its cryptographic coprocessors to secure 2 billion transactions per month.

Marriott, an early zEC12 adopter, uses the machine to centralize its reservations and streamline the selling of rooms through elastic pricing. The University of Florida turned to the z for a scalable and reliable IT infrastructure that could accommodate students accessing and interacting with applications and data through mobile applications.

Already 2013 promises to be an interesting year for z shops. Expect more innovation, optimization, and a new low cost business class machine to accompany the zEC12.

System z Clouds Pay Off

From its introduction last August, IBM has aimed the zEC12 at cloud use cases, especially private clouds. The zEC12’s massive virtualization capabilities make it possible to handle private cloud environments consisting of thousands of distributed systems running Linux on zEC12.

One zEC12, notes IBM, can encompass the capacity of an entire multi-platform data center in a single system. The newest z also enables organizations to run conventional IT workloads and private cloud applications on one system.  If you are looking at a zEC12 coupled with the zBX you can have a hybrid private cloud running Linux, Windows, and AIX workloads.

There are three main reasons why z-based data centers should consider a private cloud:

1. The z does it so naturally and seamlessly
2. It boosts IT efficiency, mainly through user self service
3. It increases enterprise agility, especially when it comes to provisioning and deploying IT resources and applications fast

Organizations everywhere are adopting private clouds (probably because C-level execs are more comfortable with private cloud security).  The Open Data Center Alliance reports faster private cloud adoption than originally predicted. Over half its survey respondents will be running more than 40% of their IT operations in private clouds by 2015.

Mainframes make a particularly good private cloud choice. Nationwide, the insurance company, consolidated 3000 distributed servers to Linux virtual servers running on a variety of z mainframes, creating a multi-platform private mainframe cloud optimized for its different workloads. The goal was to improve efficiency.

Nationwide initially intended to isolate its Linux and z/OS workloads on different physical mainframes. This resulted in a total of seven machines – a mixture of z9 and z10 servers – of which two were dedicated to Linux. To optimize this footprint, however, Nationwide ended up consolidating all workloads to four IBM zEnterprise 196 servers and two z10 servers, putting Linux and z/OS workloads on the same machines because its confidence level with Linux on the mainframe and the maturity of the platform made the Nationwide IT team comfortable mixing workloads.

The key benefit of this approach was higher utilization and better economies of scale, effectively making the mainframes into a unified private cloud—a single set of resources, managed with the same tools but optimized for a variety of workloads. The payback:  elimination of both capital and operational expenditures, expected to save about $15 million over three years. The more compact and efficient zEnterprise landscape also means low costs in the future. Specifically, Nationwide is realizing an 80% reduction in power, cooling and floor space despite an application workload that is growing 30% annually, and practically all of it handled through the provisioning of new virtual servers on the existing mainframe footprint.

Another z cloud was built by the City and County of Honolulu. It needed to increase government transparency by providing useful, timely data to its citizens. The goal was to boost citizen involvement, improve delivery of services, and increase the efficiency of city operations.

Honolulu built its cloud using an IFL engine running Linux on the city’s z10 EC machine. Between Linux and IBM z/VM the city created a customized cloud environment. This provided a scalable self-service platform on which city employees could develop open source applications, and it empowered the general public to create and deploy citizen-centric applications. Other components included IBM XIV storage, IBM Maximo Asset Management, IBM Tivoli OMEGAMON, Tivoli Workload Scheduler, and Tivoli Storage Manager.

The results: reduction in application deployment time from one week to only hours, 68% lower licensing costs for one database, and a new property tax appraisal system that increased tax revenue by $1.4 million in just three months.

There are even more examples of z clouds. For z shops a private cloud should be pretty straightforward; you’re probably over half-way there already. All you need are a few more components and a well-defined business case.  Give me a call, and I’ll even help you pull the business case together.

IBM Boosts System z Job Hunting

People looking for a mainframe job got a big boost from IBM this month with the introduction of a new System z job website called the System z Job Board. After registering for the site you get to review job postings or post one.

A quick glance at the first few pages of job posts show a range of mainframe jobs from entry-level to advanced. Companies listing System z job openings included: IBM, EMC, state of Colorado, Blue Cross/Blue Shield of Colorado, Tata, Unum, Fidelity, Humana, GT Software, state of Minnesota, CVS, and more. Even Apple was there, looking, it appeared, to fill an education and marketing position.

The zEnterprise, what amounts to a hybrid cross-platform, cross-OS enterprise server (z/OS, z/VM, Linux on z, Power, AIX, Linux, x, and soon, Windows), promises to mix up the demand curve for mainframe people and, hopefully, open new opportunities. This comes alongside the purported wave of job vacancies expected from a surge of retirement among aging mainframe veterans. That this massive wave of retirements, seemingly in defiance of demographics, has yet to materialize is puzzling. DancingDinosaur suspects the big hits many 401k plans took in recent years dampened any rosy dreams of retirement and led many to delay plans.

Momentum for the System z mainframe also continues, IBM reports, mainly in emerging markets with shops in Brazil, Mexico, Russia, China, Africa, and India having selected IBM mainframe servers in the past year, but also in North America. Payment Solution Providers (PSP), Toronto, consolidated its entire IT infrastructure on the System z after the 11-year-old company determined that an HP and Oracle infrastructure lacked the security PSP required,

 The jobs website is an extension of IBM’s Academic Initiative for System z program, a global project by IBM to align with colleges, universities and businesses across the globe to develop mainframe and large enterprise skills among college students with an eye toward future employment with Fortune 500 companies worldwide. The Academic Initiative for System z program currently involves 814 colleges and universities across the globe.

The question of mainframe skills demand has been an ongoing debate among various LinkedIn mainframe groups for years. The tenor of the debate has turned generally more positive in recent months. You can follow it here. Or join LinkedIn for free here and subscribe to any or all of the various mainframe groups.

Finally, for job hunters a number of independent mainframe people have begun to compile a worldwide registry of mainframe shops as a Wiki. When DancingDinosaur last checked nearly 500 companies were listed. You can find the list here; you will need to register if you want to submit the names of any mainframe shops not currently listed. Granted the list is incomplete; there probably are several thousand active mainframe shops in the world. Even DancingDinosaur noticed several it had covered that weren’t listed. Still, it is a terrific start; as a Wiki you can bet it will evolve, especially if you contribute.

Among unemployed mainframers looking for work, the hardest hit judging from comments posted on LinkedIn are the 50-somethings. Their problems probably have more to do with the current job market in general and attitudes toward middle-aged workers than with anything specifically to do with the mainframe.

Kookmin deploys 22 System z10 EC machines

IBM must have quite a sales rep working the financial services sector in Korea. In the last two years IBM has sold a slew of System z10 machines into the Korean financial sector alone, 29 by my count, 35 if you go back to the end of 2008.

DongBu Insurance took four System z 10 machines this past spring in a $60 million, 7-year OIO deal previously covered here. At the very end of 2009 IBM won BC Card, the leading Korea credit card processor, from HP in a deal for another three System z10 machines. Read my BC Card case study here. The Kookmin Bank deal for 22 z10 machines actually was announced in early 2009 but the bank’s IT team showed off the deployment earlier this month. A couple of months earlier (late 2008) the Industrial Bank of Korea upgraded from the z9 to six z10 EC mainframes.

Kookmin Bank is, by far, the largest in terms of the sheer number of z10 machines. They are set up in a parallel sysplex for high availability and disaster recovery. The bank, which resulted from a series of mergers over the past decade, turned to the System z consolidate all of its different business units and position the bank for future growth and change.

So why is IBM making a big deal about Kookmin Bank now? Besides the 22 z machines, a big order however you figure it, this is all about market share. As IBM noted early last year, the company nearly doubled its System z share over this past decade according to IDC’s high-end server quarterly tracker. During that time, as IBM reads the IDC numbers, HP and Sun lost share.

Since then, however, the high end server market has been tough with IBM experiencing some revenue drops.  Market shares of all vendors in the high end server market have fluxuated due to the recession, product line refreshes, and the disruption of Sun because of its acquisition by Oracle. Although Sun still is stumbling HP has been coming on strong. With IBM preparing to announce a new high end System z later this year and its recently refreshed POWER7 server lineup beginning to ramp up, expect the market share picture to continue to change.

Before the Kookmin deal in early 2009, the bank had evaluated systems and technology from IBM and other vendors, including Hewlett-Packard, Oracle and TmaxSoft. As part of its evaluation the bank also carried out benchmark tests for its core banking application and database management systems on the z and UNIX platforms.

Despite the competitive benchmarking, IBM and the z clearly had the inside track. The bank’s IT director confirmed to me that the new z10 machines were an upgrade of its existing z9 platform and did not replace any incumbent non-IBM technology. As of April 2010, the bank’s IT hardware infrastructure (including HA/DR) consisted of 22 System z10 (9-way) machines, approximately 1200 servers from IBM, HP, and Sun, and 5.5 PB of storage from IBM, EMC, and HDS.

That’s a lot of hardware, but the bank handles a heavy workload. It manages 79 million transactions daily. As for the bank’s financials: in 2009 Kookmin had assets of 256.5 trillion (won) and average operating revenue over the past three years of 31.5 trillion (won).  Note: in dollars, 10 South Korean won equal $0.01 US dollar.

The bank runs the core System z software stack: z/OS, DB2, CICS, the Rational product suite, and the Tivoli product suite. The bank does not appear to be running any non-traditional workloads on the System z via WebSphere or Linux on the z. The bank, however, operates a ton of distributed servers so it can cover those workloads outside the z.

IBM Power7 System p vs. System z

The new IBM Power7 System p looked pretty impressive at a recent briefing. So impressive that one System z staffer came up to me and asked: “What does this do to the z?”

The announcement certainly generated a lot of press, much of it focusing on the competition between IBM, HP/Intel, and Sun/Oracle in the high end UNIX market. This year is shaping up to be an enterprise server shootout.

Few seemed concerned about the implications for the System z. On the spec sheet, however, the new Power7 System p family looks potent: up to 8 cores, 32 threads, dynamic memory expansion, automated optimization, impressive SPECint benchmarks, massive software parallelization, support for AIX/Linux/System i (OS/400), and good balance between cores, memory, bandwidth, cache, and more. What’s not to like?

Throughout the briefing the System p engineers emphasized how extensively they integrated and optimized the various hardware, firmware, and software components of the new Power7 machines. Price/performance comparisons showed new optimized systems blowing away aggregations of servers from Itanium, SPARC, and x86-based rivals. IBM makes similar analyses and comes up with comparable results in comparisons between the latest z10 and various collections of HP and Sun machines.

One of the classic strengths of the mainframe is that IBM optimized, virtualized, integrated, and automated z hardware, firmware, middleware, and software from the ground up for the z. Unlike Larry Ellison vowing to optimize Sun hardware to make the most of Oracle’s software, the System z already is optimized and has been for years. Now the new System p is making a similar claim.

In a presentation last June, IBM’s Karl Freund laid out where the different IBM platforms fit in the enterprise systems scheme of things. The mainframe is designated for transaction processing and database applications based on its scalability, high transaction rates, QoS, ability to handle peak workloads, resiliency, and security. IBM’s UNIX platforms like System p are designated for business applications, analytics, and HPC.

OK, I’ll grant HPC should go to UNIX, but business apps and analytics are arguable. Why put WebSphere on z if not for business applications or Cognos for Linux on z if not for BI and analytics? The plain truth: the z and p platforms compete for some workloads.

When asked about the overlap between the z and p platforms IBM executives see no problem: the Power7 System p is for UNIX applications. OK, System z doesn’t run AIX and Power7 System p doesn’t run z/OS, but both can run Linux.

This may become a moot issue when a new System z arrives, as expected, later this year. That’s the hybrid z discussed here last week. It promises, in IBM-speak, to simplify, consolidate, and reduce the costs of managing IT infrastructure by integrating, virtualizing, and coherently managing the multiple and varied heterogeneous processing elements of a deployed business service.

Over the past few months I’ve been speaking with System p users to better understand its appeal. For many that means AIX, which you won’t find on the z. Still, the System z and p stories could become quite entangled this year, especially when you throw in the platform moves of IBM’s competitors. Watch this space as things develop.

 

TmaxSoft lowers IBM z/OS computing cost

Even as IBM battles NEON in court in an effort to squelch zPrime, another upstart is emerging on the mainframe scene, TmaxSoft, which is about to introduce a System z version of OpenFrame, promises to duplicate key z/OS functionality, such as CICS, VSAM, RACF, JES and more on a low cost IFL.

TmaxSoft, in effect, offers another way to reduce the cost of System z computing by avoiding mainframe software licensing costs. Instead you run these traditional mainframe applications on an IFL. And as with zPrime, big, big savings here. eWeek reported it here, but it didn’t seem to get picked up much.

As the Register, a UK publication, noted in January, IBM can’t let this get out of hand: “With mainframe revenues off sharply and likely to be so until the System z11 mainframes ship much later this year, IBM can ill afford to look the other way as Neon Software peddles its zPrime tool for offloading mainframe workloads to much cheaper specialty engines.” So it sued Neon, which had already launched its own pre-emptive legal strike at the end of 2009.

TmaxSoft hopes to avoid a similar confrontation with IBM. “We’re a different things. NEON encouraged users to violate their license [according to IBM]. They couldn’t come after us with the same claim. Our customers take their own intellectual property and run it on OpenFrame. There is no IBM software involved,” says John Plato, the TmaxSoft marketing person in the US.

Still, TmaxSoft offers the opportunity for what Plato describes as workload shifting. OpenFrame duplicates the functionality of core z/OS systems. replacing the legacy CICS, IMS, JES, VSAM mainframe engines with OpenFrame to run the customer’s applications originally written in legacy code like COBOL and PL1 on Linux, but do so without having to make changes to the source code or business logic. It also handles batch applications.

This allows companies to move those applications to the IFL, which reduces the licensing costs without having to give up the benefits of the mainframe, just the expense of running software under z/OS. The result is the same as with zPrime; the customer runs what had been a z/OS application on an IFL, at the substantially lower IFL licensing costs.

TmaxSoft, in effect, emulates those core z/OS functions in OpenFrame although it doesn’t call it emulation. And the performance is as good as the real thing based on a test of a 7500 MIPS workload, according to Plato. The product is ready for production use, he reports; the company is just looking to sign up some pilot projects.

Whether TmaxSoft avoids IBM’s legal wrath remains to be seen. As the Register reported, IBM’s charge is that NEON is hijacking IBM intellectual property. TmaxSoft is hoping to avoid that by inserting its own intellectual property instead. The lawyers will have to figure this one out.

The real point, however, is that mainframe computing is perceived as simply too expensive, especially with the availability of powerful, less costly platforms that promise to handle mainframe-class workloads without the mainframe overhead. Whether they actually deliver on that promise is another question. But one thing is certain: upstarts like NEON and TmaxSoft will keep coming as long as there is a demand for lower cost mainframe computing.