Everyone knows that mainframes are the best computers you can have. You can run them locally, you can hide them in the cloud, and you can link them together into a massive processing network. But we also know that there are smaller platforms out there that work differently. Wouldn’t it be brilliant if you could run them all from one place?
Last summer we were excited by the announcement from IBM of its new zBC12 mainframe computer. The zBC12 followed the previous year’s announcement of the zEC12 (Enterprise Class), and 2011 saw the z114, with 2010 giving us the z196. So what’s special about those mainframes?
Well, in addition to IFL, zIIP, and zAAP specialty processors, and massive amounts of processing power, they came with the IBM zEnterprise BladeCenter Extension (zBX), which lets users combine workloads designed for mainframes with those for POWER7 and x86 chips, like Microsoft Windows Server. So let’s unpick this a little.
One issue many companies have after years of mergers and acquisitions is a mixed bag of operating systems and platforms. They could well have server rooms across the world and not really know what was running on those servers.
IBM’s first solution to this problem was Linux on System z. Basically, a company could take hundreds of Linux servers and consolidate them onto a single mainframe. They would save on energy to drive the servers and cool them, and they would get control of their IT back.
IBM’s second solution, as we’ve just described, was to incorporate the hardware for Linux and Windows servers in its mainframe boxes. You’d just plug in the blades that you needed and you had full control over your Windows servers again (plus all the benefits of having a mainframe).
But what about if you could actually run Windows on your mainframe? That was the dream of some of the people at Mantissa Corporation. They did a technology demo at SHARE in August 2009. According to Mantissa’s Jim Porell, Network World read their abstract and incorrectly assumed that they were announcing a product – which they weren’t. The code is still in beta. But think about what it could mean: running all your Windows servers on a mainframe. That is quite a concept.
Again, according to Jim, they can now have real operating systems running under their z86VM, although, so far, they are the free versions of Linux. Their next step will be to test it with industrial strength Linux distros such as Red Hat and Suse. And then, they will need to get Windows running. And then they’ll have a product.
Speaking frankly, Jim said that currently they have a bug in their Windows ‘BIOS’ processing in the area of plug-and-play hardware support. Their thinking is that it’s a mistake in their interpretation of the hardware commands and, naturally, they’re working to resolve it.
The truth is that it’s still early days for the project, and while running Linux is pretty good, we can already do that on a mainframe (although you might quibble at the price tag for doing so). But once the Mantissa technical people have cracked the problems with Windows, it will be a product well worth taking a look at, I’m sure. But they’re not there yet, and they’re keen to manage expectations appropriately.
Jim goes on to say that once the problems are solved it’s going to be about performance. Performance is measured in a couple of ways: benchmarks of primitives, benchmarks of large-scale for capacity planning, end user experiences, and what they are willing to tolerate. So it seems that they have business objectives around performance where they could be successful if they supported only 10 PCs, then more successful with a 100 PCs, and even have even greater success if they can support a 1000 PCs.
Jim Porell describes z86VM as really just an enabler to solving a wide range of ‘customer problems’ by enabling direct access between the traditional mainframe and the PC operating systems that are co-located with it.
I think that anything that gets more operating systems running on mainframe hardware has got to be good. And I’m prepared to wait for however long it takes Mantissa to get Windows supported on a mainframe. I’ll definitely feel then that I’m having my cake and eating it!
Showing posts with label zBX. Show all posts
Showing posts with label zBX. Show all posts
Sunday, 15 June 2014
Sunday, 30 October 2011
Two things you thought would never happen at IBM
I guess any two pundits sitting in a room together 10 years ago and talking about IBM’s future would have been more likely to predict Star Trek-like beaming technology and computers you could talk to than a mainframe that integrated Windows servers and woman landing the top job at IBM.
And here we are. It’s almost November 2011, and both are about to come to pass.
The zEnterprise 196 and the Business Class version, the zEnterprise 114, mainframes come with the zEnterprise BladeCenter Extension. Initially this supported AIX on Power blades and Linux on x86 blades. This fit nicely with IBM’s model of the universe because it owns AIX and Linux is, of course, open source – ie it doesn’t belong to anybody. The Unified Resource Manager (URM) controls the operating systems and hypervisors on the mainframe and the blades. But now – the previously unthinkable – IBM promises that it will have Windows running on its HX5 Xeon-based blade servers for the zBX chassis before the end of this year.
Microsoft Windows Server 2008 R2 Datacenter Edition will run on the PS701 blade servers in the zBX enclosures. The zBX extension can have 112 PS701 blades or 28 HX5 blades.
This is clearly important for those sites that use mainframes or are ready to upgrade to mainframes and still have a big Windows-using population. It’s interesting that so many people consider Windows to be the de facto computing platform. I recently had a conversation where Windows laptops were given the metaphor of rats or beetles – they just turn up everywhere – and Linux was given the metaphor of a stealth operating system or a hidden shadow – it was everywhere, but you didn’t see it. Why stealth, well because Linux turns up behind the scenes on routers, on TiVO boxes, on supercomputers, as the precursor to Android on smartphones, making movies at Pixar and Dreamworks, in the military, governments, everywhere!
After Windows on IBM hardware, the next thing we hear is that Virginia M Rometty, a senior vice president at IBM, is going to be the company’s next CEO – starting in January. “Ginni”, aged 54 (as all the releases inform us), succeeds Samuel J Palmisano, who is 60, and will remain as chairman.
Ms Rometty graduated from Northwestern University with a degree in computer science, joined IBM in 1981 as a systems engineer. She moved through different management jobs, working with clients in a variety of industries. Her big coup was in 2002, when she played a major part in the purchase of the very big consulting firm, PricewaterhouseCoopers Consulting. PwC staff were used to working in a different way from IBM’s and managing that culture shift was down to Ms Rometty.
In 2009, Ginni became senior vice president and group executive for sales, marketing, and strategy.
You’ll recall that Sam Palmisano took over in 2003 from Louis V Gerstner Jr, who’d joined IBM from RJR Nabisco in 1993 and helped turn round an ailing IBM. The previous incumbent had been the lacklustre John Akers.
I suppose with Siri on iPhones and the much less serious about itself Iris on Android, we’ve moved some way towards being able to talk to a computer – even if it is a smartphone. Still no sign of Scotty being beamed up, though!
And here we are. It’s almost November 2011, and both are about to come to pass.
The zEnterprise 196 and the Business Class version, the zEnterprise 114, mainframes come with the zEnterprise BladeCenter Extension. Initially this supported AIX on Power blades and Linux on x86 blades. This fit nicely with IBM’s model of the universe because it owns AIX and Linux is, of course, open source – ie it doesn’t belong to anybody. The Unified Resource Manager (URM) controls the operating systems and hypervisors on the mainframe and the blades. But now – the previously unthinkable – IBM promises that it will have Windows running on its HX5 Xeon-based blade servers for the zBX chassis before the end of this year.
Microsoft Windows Server 2008 R2 Datacenter Edition will run on the PS701 blade servers in the zBX enclosures. The zBX extension can have 112 PS701 blades or 28 HX5 blades.
This is clearly important for those sites that use mainframes or are ready to upgrade to mainframes and still have a big Windows-using population. It’s interesting that so many people consider Windows to be the de facto computing platform. I recently had a conversation where Windows laptops were given the metaphor of rats or beetles – they just turn up everywhere – and Linux was given the metaphor of a stealth operating system or a hidden shadow – it was everywhere, but you didn’t see it. Why stealth, well because Linux turns up behind the scenes on routers, on TiVO boxes, on supercomputers, as the precursor to Android on smartphones, making movies at Pixar and Dreamworks, in the military, governments, everywhere!
After Windows on IBM hardware, the next thing we hear is that Virginia M Rometty, a senior vice president at IBM, is going to be the company’s next CEO – starting in January. “Ginni”, aged 54 (as all the releases inform us), succeeds Samuel J Palmisano, who is 60, and will remain as chairman.
Ms Rometty graduated from Northwestern University with a degree in computer science, joined IBM in 1981 as a systems engineer. She moved through different management jobs, working with clients in a variety of industries. Her big coup was in 2002, when she played a major part in the purchase of the very big consulting firm, PricewaterhouseCoopers Consulting. PwC staff were used to working in a different way from IBM’s and managing that culture shift was down to Ms Rometty.
In 2009, Ginni became senior vice president and group executive for sales, marketing, and strategy.
You’ll recall that Sam Palmisano took over in 2003 from Louis V Gerstner Jr, who’d joined IBM from RJR Nabisco in 1993 and helped turn round an ailing IBM. The previous incumbent had been the lacklustre John Akers.
I suppose with Siri on iPhones and the much less serious about itself Iris on Android, we’ve moved some way towards being able to talk to a computer – even if it is a smartphone. Still no sign of Scotty being beamed up, though!
Sunday, 18 September 2011
Mainframe maintenance – a new paradigm with new challenges
For many organizations, we’re beginning to see a new model of how IT customer support can be organized – and the model is coming from management who are completely platform-agnostic. To them, IT is IT – it doesn’t matter whether something runs on a mainframe or a distributed platform. And this new way of working brings with it new challenges.
This whole change in staff structure is also being encouraged by the advent of the zEnterprise hybrid machines with their zBX blades running everything from AIX to, potentially, Windows. A consequence is that a mainframe specialist could be dealing with a Linux error message, or a Windows SharePoint guru might be trying to understand what’s going on inside CICS. What can you do to help them?
Or let’s suppose in a more traditional mainframe environment, for whatever reason, you lost some of your top technical people. Perhaps they got jobs elsewhere or perhaps they retired early, but suddenly you find yourself with a huge knowledge gap. Maybe you can transfer someone across from the distributed team. Or maybe you can recruit one of the new generation of youngsters who are learning the benefits of mainframe computing. But whatever you do, there will be a fairly long period of time during which anything out of the ordinary occurring is going to leave everyone scratching their heads and searching Google – whereas, previously, your in-house expert knew exactly what to do. So, in this situation, what are you going to do?
Let’s not worry too much at this stage about Service Level Agreements (SLAs) and performance targets. Let’s simply focus on the problem. How can any organization, irrespective of how its IT team is constructed, ensure that appropriate expertise is available at all times to whichever staff members are available?
Obviously you can have the manuals, and some could be on the IBM mainframe portal, but that doesn’t give you speedy access to the necessary information. A Google search will reveal hundredsof pages of results, but it takes a degree of expertise to sift through those and find the correct one quickly. And someone without any expertise could spend a very long time reading solutions to completely different problems before ever finding the right one. Not a satisfactory way to provide IT services to customers – whether internal or external to the organization.
So what would be a good solution? How can these issues of staff working outside their comfort zone be dealt with in a way that is good for the business? And what kind of a solution will still be able to ensure those business-critical mainframes are being supported in a year’s time, in five year’s, or even further into the future?
This is where a new breed of solutions that can address this coming challenge are positioning themselves. One of these, Softlib with its iSolve product (www.softlibsw.com/mvs.aspx), allows an organization to combine all its IT-related information into a single virtual library. That means users – your harassed staff – have to search in only one place, not as previously in many places, to find a solution to any problem. And once you have a single location for information available, you can allow product champions and other IT-literate staff access to it – which should result in more empowered and satisfied users and fewer calls to the Help Desk.
It makes sense to organize the information in this single virtual library using themes, so CICS information might be one theme, IMS another, Linux a third, etc. The information in the library starts from IBM and third-party software vendors’ manuals, and can be supplemented with information from newsgroups and other online resources. Plus, you can add your own technical expertise.
Access to the information can be from a Web browser or a terminal server. It can be hosted locally, or as a cloud-based resource. The advantage of the cloud route is that the information is looked after by Softlib and they already have access to a huge number of the resources you’ll need. So you can start using the facility almost immediately. Plus the online documentation is automatically updated when new information becomes available. Other benefits include knowledge usage analytics that can help address missing or outdated knowledge, and seamless integration with CRM, bug-tracking, Service Desk, content-management applications, etc.
All in all, Softlib’s iSolve product has a lot to offer most mainframe sites, and certainly provides an answer to the question of what to do if you restructure your IT customer support and need to extend the working expertise of your staff onto other platforms such as AIX and Windows. It also offers a solution to the problem of losing key mainframe experts in a mainframe-only environment.
This whole change in staff structure is also being encouraged by the advent of the zEnterprise hybrid machines with their zBX blades running everything from AIX to, potentially, Windows. A consequence is that a mainframe specialist could be dealing with a Linux error message, or a Windows SharePoint guru might be trying to understand what’s going on inside CICS. What can you do to help them?
Or let’s suppose in a more traditional mainframe environment, for whatever reason, you lost some of your top technical people. Perhaps they got jobs elsewhere or perhaps they retired early, but suddenly you find yourself with a huge knowledge gap. Maybe you can transfer someone across from the distributed team. Or maybe you can recruit one of the new generation of youngsters who are learning the benefits of mainframe computing. But whatever you do, there will be a fairly long period of time during which anything out of the ordinary occurring is going to leave everyone scratching their heads and searching Google – whereas, previously, your in-house expert knew exactly what to do. So, in this situation, what are you going to do?
Let’s not worry too much at this stage about Service Level Agreements (SLAs) and performance targets. Let’s simply focus on the problem. How can any organization, irrespective of how its IT team is constructed, ensure that appropriate expertise is available at all times to whichever staff members are available?
Obviously you can have the manuals, and some could be on the IBM mainframe portal, but that doesn’t give you speedy access to the necessary information. A Google search will reveal hundredsof pages of results, but it takes a degree of expertise to sift through those and find the correct one quickly. And someone without any expertise could spend a very long time reading solutions to completely different problems before ever finding the right one. Not a satisfactory way to provide IT services to customers – whether internal or external to the organization.
So what would be a good solution? How can these issues of staff working outside their comfort zone be dealt with in a way that is good for the business? And what kind of a solution will still be able to ensure those business-critical mainframes are being supported in a year’s time, in five year’s, or even further into the future?
This is where a new breed of solutions that can address this coming challenge are positioning themselves. One of these, Softlib with its iSolve product (www.softlibsw.com/mvs.aspx), allows an organization to combine all its IT-related information into a single virtual library. That means users – your harassed staff – have to search in only one place, not as previously in many places, to find a solution to any problem. And once you have a single location for information available, you can allow product champions and other IT-literate staff access to it – which should result in more empowered and satisfied users and fewer calls to the Help Desk.
It makes sense to organize the information in this single virtual library using themes, so CICS information might be one theme, IMS another, Linux a third, etc. The information in the library starts from IBM and third-party software vendors’ manuals, and can be supplemented with information from newsgroups and other online resources. Plus, you can add your own technical expertise.
Access to the information can be from a Web browser or a terminal server. It can be hosted locally, or as a cloud-based resource. The advantage of the cloud route is that the information is looked after by Softlib and they already have access to a huge number of the resources you’ll need. So you can start using the facility almost immediately. Plus the online documentation is automatically updated when new information becomes available. Other benefits include knowledge usage analytics that can help address missing or outdated knowledge, and seamless integration with CRM, bug-tracking, Service Desk, content-management applications, etc.
All in all, Softlib’s iSolve product has a lot to offer most mainframe sites, and certainly provides an answer to the question of what to do if you restructure your IT customer support and need to extend the working expertise of your staff onto other platforms such as AIX and Windows. It also offers a solution to the problem of losing key mainframe experts in a mainframe-only environment.
Sunday, 17 July 2011
Command economies, decentralization, and the z114
It comes and goes. It’s like a pendulum swinging in one direction, running out of steam, and then swinging in the completely opposite direction. And it applies to countries, economies, and the way people view computing. Let me explain...
During the 1970s, computing, where it existed, was very much a centralized affair. The gods of the mainframe pretty much controlled what anyone was able to do. It was like Stalinist Russia. Everything came out of the centre. You didn’t get it, unless someone at the hub of things deemed it necessary for you to have it.
Currently in the UK, we have the opposite approach in terms of our model of how things should work. Quite logically, you might think, if you live in a rural area with rolling fields full of wheat or livestock, your concerns are completely different from those of someone living in a post-industrial run-down urban area. Of course, this localism easily lends itself to the criticism of postcode lottery. Anyway, we have little islands of individuality separate from each other. Unfortunately, the reality is that political areas tend to include more than a monoculture of just rural or just urban populations. Plus you have different needs for different age groups – you can see where this idea falls down when applied to the real world, but hang on to the little islands metaphor.
Now let’s turn time back to 1989. We find the Berlin wall coming down and the whole centralized power base of the USSR and it’s Warsaw Pack allies crumbling. In the world of computing, we find the balance of opinion has moved right away from mainframes. In the early 90s, their death was confidently predicted. In its place we had millions of underpowered PCs running DOS-based operating systems. And as the 90s progressed we saw the triumph of Windows and Microsoft. We also saw that antithesis of centralization, Open Source software. Unix started life in 1969, and Linus Torvalds’ Linux arrived in 1991. Even IBM, which had developed and standardized the PC in the early 1980s, was working on the development of other platforms. 1988 gave us the AS/400 – now the IBM System i and which now runs on the POWER platform. The RS/6000, running a Unix variant called AIX, arrived in the 1990s and also now run on POWER hardware.
So, having been empowered to make their own decisions and choices of hardware and software, what have users done since then? Well, in the PC world, they go for big servers that are virtualized in order to benefit from the control that gives them. It makes back-ups and business continuity easier.
And now, here we are in 2011 and IBM announces a Business Class (basically not a top-end machine, more one for the everyman mainframe user) zEnterprise – the z114. It’s gone back to being a centralized piece of hardware because not only is it a mainframe, it’s a POWER7 box, and it has x86 blades. So that gives users a smaller footprint, less power consumption, and control of everything using the IBM zEnterprise Unified Resource Manager and the IBM zEnterprise BladeCenter Extension (zBX). The POWER7 blades mean that AS/400 and RS/6000-heritage users have a home. And the x86 blades not only run Linux x86 applications unchanged, but, by the end of the year, are expected to run Windows applications too.
The culture shock at many sites will come when the distributed applications teams (and they may have many different names) discover that all the things they’ve been planning to achieve (virtualized desktops, virtualized servers, etc) are just part of the techniques that the mainframe people take for granted. And the mainframers are going to have to understand that for many of the people in the other teams, it’s in many ways still about 1992 in terms of business recovery times etc. But when the teams do come together, the synergy is going to be very beneficial for the organization that allow it to happen.
This new mainframe, unusually, comes with a published price tag – $75,000. As part of the package you get the IBM Smart Analytics Optimizer to analyse data faster at a lower cost per transaction, and the IBM WebSphere DataPower XI50 for integrating Web-based workloads. The new hardware runs the latest version of z/OS – 1.13. You get 3.8GHz processors (the zEnterprise 196 uses 5.3GHz processors), and you can configure up to 14 of them with 10 specialty processors – zIIP, zAAP, and IFL.
The pendulum has now swung completely back. We have a single box capable of providing all the different computing needs of an organization.
During the 1970s, computing, where it existed, was very much a centralized affair. The gods of the mainframe pretty much controlled what anyone was able to do. It was like Stalinist Russia. Everything came out of the centre. You didn’t get it, unless someone at the hub of things deemed it necessary for you to have it.
Currently in the UK, we have the opposite approach in terms of our model of how things should work. Quite logically, you might think, if you live in a rural area with rolling fields full of wheat or livestock, your concerns are completely different from those of someone living in a post-industrial run-down urban area. Of course, this localism easily lends itself to the criticism of postcode lottery. Anyway, we have little islands of individuality separate from each other. Unfortunately, the reality is that political areas tend to include more than a monoculture of just rural or just urban populations. Plus you have different needs for different age groups – you can see where this idea falls down when applied to the real world, but hang on to the little islands metaphor.
Now let’s turn time back to 1989. We find the Berlin wall coming down and the whole centralized power base of the USSR and it’s Warsaw Pack allies crumbling. In the world of computing, we find the balance of opinion has moved right away from mainframes. In the early 90s, their death was confidently predicted. In its place we had millions of underpowered PCs running DOS-based operating systems. And as the 90s progressed we saw the triumph of Windows and Microsoft. We also saw that antithesis of centralization, Open Source software. Unix started life in 1969, and Linus Torvalds’ Linux arrived in 1991. Even IBM, which had developed and standardized the PC in the early 1980s, was working on the development of other platforms. 1988 gave us the AS/400 – now the IBM System i and which now runs on the POWER platform. The RS/6000, running a Unix variant called AIX, arrived in the 1990s and also now run on POWER hardware.
So, having been empowered to make their own decisions and choices of hardware and software, what have users done since then? Well, in the PC world, they go for big servers that are virtualized in order to benefit from the control that gives them. It makes back-ups and business continuity easier.
And now, here we are in 2011 and IBM announces a Business Class (basically not a top-end machine, more one for the everyman mainframe user) zEnterprise – the z114. It’s gone back to being a centralized piece of hardware because not only is it a mainframe, it’s a POWER7 box, and it has x86 blades. So that gives users a smaller footprint, less power consumption, and control of everything using the IBM zEnterprise Unified Resource Manager and the IBM zEnterprise BladeCenter Extension (zBX). The POWER7 blades mean that AS/400 and RS/6000-heritage users have a home. And the x86 blades not only run Linux x86 applications unchanged, but, by the end of the year, are expected to run Windows applications too.
The culture shock at many sites will come when the distributed applications teams (and they may have many different names) discover that all the things they’ve been planning to achieve (virtualized desktops, virtualized servers, etc) are just part of the techniques that the mainframe people take for granted. And the mainframers are going to have to understand that for many of the people in the other teams, it’s in many ways still about 1992 in terms of business recovery times etc. But when the teams do come together, the synergy is going to be very beneficial for the organization that allow it to happen.
This new mainframe, unusually, comes with a published price tag – $75,000. As part of the package you get the IBM Smart Analytics Optimizer to analyse data faster at a lower cost per transaction, and the IBM WebSphere DataPower XI50 for integrating Web-based workloads. The new hardware runs the latest version of z/OS – 1.13. You get 3.8GHz processors (the zEnterprise 196 uses 5.3GHz processors), and you can configure up to 14 of them with 10 specialty processors – zIIP, zAAP, and IFL.
The pendulum has now swung completely back. We have a single box capable of providing all the different computing needs of an organization.
Friday, 23 July 2010
zEnterprise - three into one does go!
On Thursday 22 July 2010, IBM announced not just their latest mainframe – the zEnterprise – but also a roadmap of how data centres should be – unified, in a single box, and controlled by the best equipped platform in that box. Once code-named Gryphon (you know, the mythical creature with an eagle’s head and lion’s body), IBM’s new hybrid machine (although in the UK you might hear it referred to as a coalition!) brings together the latest mainframe technology with POWER7 and x86 IBM blade systems, giving potential users z/OS, AIX, and Linux all on the one box. And all this is controlled from the mainframe console by the new Unified Resource Manager. You might like to think of this new mainframe as a virtualization hub that manages other workloads in the data centre. Three platforms in one box.
IBM’s view is that data centres are running more than one lot of hardware, and sites are experiencing problems with space for the hardware, keeping control of these different systems, and even communicating between them – so integrating them seems like the obvious answer. And that’s what they’ve done.
But more has come out of the $1.5 billion spent on research and development. The zEnterprise 196 (IBM's name for the two-rack mainframe) includes 96 5.2GHz (up from 4.4GHz on the z10) quad processors (80 of which are used by the client, and the rest are used by the machine itself) and up to 3TB of memory (double the z10s). The new microprocessors offer 100 new mainframe machine code instructions.
In terms of performance, the zEnterprise can handle 50 billion instructions per second, providing a 40-60% performance increase over the z10 without using any more power. A water cooling option, could help reduce energy consumption by up to 12 percent by removing air heat. It also includes the first implementation of RAID memory, which is like RAID for disks, and could be used to increase uptime to beyond the 99.999 availability of current mainframe technology.
The zEnterprise BladeCenter Extension (zBX) operates as a tightly-coupled extension to the mainframe through a high-performance private network. The users then add POWER7 or System x blades to four racks. IBM says that POWER7 blades will be available by the end of this year, with System x blades scheduled to ship in the first half of 2011. The new Unified Resource Manager will allow users to install, monitor, manage, optimize, diagnose, and service resources and workloads from a single console across the entire infrastructure.
The new machine includes a DB2 accelerator, called the Smart Analytics Optimzer, which is able to routes database queries either to the mainframe DB2 system or a specialist blade server optimized for smart analytics. IBM estimates that complex database queries can experience up to a ten-fold performance improvement in this environment.
Can you afford it? IBM suggests using zEnterprise will reduce cost of ownership by 55% – but didn’t disclose the actual price. They did say that its price relative to capacity would be lower than for the z10!
IBM’s view is that data centres are running more than one lot of hardware, and sites are experiencing problems with space for the hardware, keeping control of these different systems, and even communicating between them – so integrating them seems like the obvious answer. And that’s what they’ve done.
But more has come out of the $1.5 billion spent on research and development. The zEnterprise 196 (IBM's name for the two-rack mainframe) includes 96 5.2GHz (up from 4.4GHz on the z10) quad processors (80 of which are used by the client, and the rest are used by the machine itself) and up to 3TB of memory (double the z10s). The new microprocessors offer 100 new mainframe machine code instructions.
In terms of performance, the zEnterprise can handle 50 billion instructions per second, providing a 40-60% performance increase over the z10 without using any more power. A water cooling option, could help reduce energy consumption by up to 12 percent by removing air heat. It also includes the first implementation of RAID memory, which is like RAID for disks, and could be used to increase uptime to beyond the 99.999 availability of current mainframe technology.
The zEnterprise BladeCenter Extension (zBX) operates as a tightly-coupled extension to the mainframe through a high-performance private network. The users then add POWER7 or System x blades to four racks. IBM says that POWER7 blades will be available by the end of this year, with System x blades scheduled to ship in the first half of 2011. The new Unified Resource Manager will allow users to install, monitor, manage, optimize, diagnose, and service resources and workloads from a single console across the entire infrastructure.
The new machine includes a DB2 accelerator, called the Smart Analytics Optimzer, which is able to routes database queries either to the mainframe DB2 system or a specialist blade server optimized for smart analytics. IBM estimates that complex database queries can experience up to a ten-fold performance improvement in this environment.
Can you afford it? IBM suggests using zEnterprise will reduce cost of ownership by 55% – but didn’t disclose the actual price. They did say that its price relative to capacity would be lower than for the z10!
Subscribe to:
Posts (Atom)