Feeds

IBM preps mini-mainframe for launch

The System z upgrade cycle continues

Internet Security Threat Report 2014

The word on the street is that Big Blue is getting set to launch the so-called Business Class iteration of its System z mainframe, a midrange-class machine to complement last year's System zEnterprise 196 server. The announcement is set for some time in July, most likely in the first week after the 4th of July holiday in the United States.

The timing for the new BC mainframe is about right. The System zEnterprise 196 machines made their debut last July and were a big part of IBM's recovery in server sales for the last quarter of 2010 and the first quarter of 2011. In fact, the zEnterprise 196 machines, which are based on the quad-core z11 processor, generated more than $1bn in revenues for base systems alone in each of those two quarters, and are expected to sell at a pretty healthy clip for the remainder of the year. The BC-class machines are generally less scalable, have slower processors, and lower price tags and are therefore more appealing to midrange shops with z/VSE, z/VM, and z/OS workloads.

The zEnterprise 196 mainframes, also known as the Enterprise Class or EC line, are based on a 5.2GHz, four-core CISC processor designed and fabbed by IBM in its Poughkeepsie, New York, mainframe stomping grounds. The z196, as it is sometimes called, puts six of these four-core z11 processors onto a single system board – what IBM calls a processor book in the System z and high-end Power Systems lines. In the case of the z196, up to four of these books can be configured into a single system image with as many as 80 cores dedicated to running a single workload across up to 3TB of DDR3 main memory. That main memory has data striping, like RAID disk arrays do.

Each one of those z11 cores is rated at around 1,200 MIPS running at 5.2GHz, and a system with all 80 cores activated to run z/OS workloads can deliver an aggregate of 52,000 MIPS. IBM offers five different models of the z196 – the M15, M32, M49, M66, and M80 – with the number designating the maximum number of central processors (CPs in the mainframe lingo) that can be activated from the core pool inside the boxes. As with prior zSeries and System z mainframes, the z196 can also have cores configured as specialty processors for supporting Linux (IFL) or for accelerating Java and XML processing (zAAP) or DB2 databases (zIIP). The cores can also be configured as coupling facilities (CFs) to link multiple machines into a Parallel Sysplex cluster.

The System z10 BC machine came out in October 2008, offered up to five CP engines and then up to 10 IFL Linux engines or 10 sysplex coupling facilities; five engines can be designated as zIIPs or zAAPs as well. The BC machines are popular for customers with modest COBOL applications running on IBM's z/OS operating system – what Big Blue internally calls Classic Mainframe, in deference to Coca-Cola – but are also popular among a minority of customers who like the openness and relative low cost of Linux but like the reliability and security of mainframes – what IBMers call New Mainframe.

While the z10 engines used in the quad-core System z10 EC mainframes ran at 4.4GHz, the chips used in the System z10 BC models announced almost three years ago were geared down to 3.5GHz and one of the cores on each chip was a dud that was isolated. IBM originally supported from 4GB to 120GB of DDR2 main memory on the z10 BC machines, and in June 2009 boosted that to 248GB.

If history is any guide, the System zEnterprise BC machines based on the z11 processors will follow a similar trend, with one or more cores on the die being an electrically isolated dud. If clock speeds scale as they did in the z10 generation, then the zEnterprise BC machine should have cores running at around 4.1GHz, delivering something on the order of 980 MIPS of processing capacity per engine.

If IBM keeps the zEnterprise BC midrange mainframe at five CPs, then a single system image should be able to span up to 3,400 MIPS or so. The zEnterprise 196 big boy has 3TB across 80 cores, or 38.4GB per core, which is a lower ratio already supported by the z10 BC box from three years ago, which sported 24GB per core at launch in late 2008 and 49.6GB per core the following summer. IBM may be tempted to keep the memory capacity on the zEnterprise BC box more or less the same.

Ditto for the core count. It seems unlikely that IBM will increase the number of CP cores in the new BC box because to do so would eat into the low-end of zEnterprise 196 EC sales, and this would be bad because BC machines, MIPS for MIPS, are a lot less expensive than EC machines. The original BC machines from 2006 had either a single core that could be used as a CP, or from 1 to 3 cores or from 1 to 4 cores, depending on the model. The follow-on BC machines – the ones due to be replaced – had 5 CPs.

It will be interesting to see if IBM adds the zBX blade server extensions to the zEnterprise BC models that started shipping last year for the EC machines. With the zBX configuration, IBM is allowing Power and Xeon blade servers to run AIX, Linux, and eventually Windows workloads in a fashion that is tightly coupled to the mainframe and on internal networks that are secure from the outside world.

The last thing is, what will this BC mainframe be called? IBM's marketeers are insane, but there is a logic to what they do if you look closely. The zEnterprise 196 EC is a big mainframe with 96 cores total in the box, but only 80 available for the operating system. So to be consistent, if IBM supports five CP engines in the BC variant, but there are ten total engines, then it should be called the zEnterprise 110 BC. No matter what, these are terrible names. There was nothing wrong at all with System z11, except that some marketer has to muck about with this stuff to justify his or her existence. ®

Internet Security Threat Report 2014

More from The Register

next story
NSA SOURCE CODE LEAK: Information slurp tools to appear online
Now you can run your own intelligence agency
Azure TITSUP caused by INFINITE LOOP
Fat fingered geo-block kept Aussies in the dark
NASA launches new climate model at SC14
75 days of supercomputing later ...
Yahoo! blames! MONSTER! email! OUTAGE! on! CUT! CABLE! bungle!
Weekend woe for BT as telco struggles to restore service
Cloud unicorns are extinct so DiData cloud mess was YOUR fault
Applications need to be built to handle TITSUP incidents
BOFH: WHERE did this 'fax-enabled' printer UPGRADE come from?
Don't worry about that cable, it's part of the config
Stop the IoT revolution! We need to figure out packet sizes first
Researchers test 802.15.4 and find we know nuh-think! about large scale sensor network ops
SanDisk vows: We'll have a 16TB SSD WHOPPER by 2016
Flash WORM has a serious use for archived photos and videos
Astro-boffins start opening universe simulation data
Got a supercomputer? Want to simulate a universe? Here you go
prev story

Whitepapers

Go beyond APM with real-time IT operations analytics
How IT operations teams can harness the wealth of wire data already flowing through their environment for real-time operational intelligence.
10 threats to successful enterprise endpoint backup
10 threats to a successful backup including issues with BYOD, slow backups and ineffective security.
Forging a new future with identity relationship management
Learn about ForgeRock's next generation IRM platform and how it is designed to empower CEOS's and enterprises to engage with consumers.
High Performance for All
While HPC is not new, it has traditionally been seen as a specialist area – is it now geared up to meet more mainstream requirements?
Website security in corporate America
Find out how you rank among other IT managers testing your website's vulnerabilities.