Mainframe Linux


Overview

So you’re working with SLES or SLED, and someone asks you if a particular package or product is supported. What do you do? Of all the options available, I refer constantly to the following pages and references to make sure I give accurate and helpful information to the querying person.

Where to Go

The top page I usually refer people to in this case is the Tech Specs section. On this page is listed SLES, SLED and Partner hardware and software support links. If I click on the “SuSE Linux Enterprise Server” link, I will reach the technical specifications page for SuSE Linux Enterprise Server. We’ve seen this page in other posts, such as How Many ____’s Does SLES Support? – Part I, where we discussed Kernel Limits and File System Features.

The chief thing I want to determine when someone asks me about a particular package’s support is that we do include it in our distributions. I query the person about what version and patch level they want to know about, and then click on the appropriate link, which is listed in the paragraph below the menu tabs, “see the list of supported packages”.

This link leads me to the “SuSE Linux Enterprise Server Supported Packages page, where for example I will pick the link for “SuSE Linux Enterprise Server 10 SP1 for X86“, a PDF file that I can then search for a package by pressing Ctrl-f, and typing the name I’m looking for, either in my browser, or my PDF reader that popped up.

All Together Now

My example would be someone asking me if there is support for Mono (a Dot Net compatible server solution) on SLES 10 SP1, both on the X86 and System Z platforms.

Those PDF files will tell me very quickly that there is support all the way to Level 3 for SLES 10 SP1 on X86 and the IBM System Z Mainframe.

Hopefully this will help you determine quickly what packages are included in the various distributions and patch levels, and make it easier to find what level of support is provided for those packages.

Enjoy,

RossB

From the article:

CHICAGO — At the Data Center Decisions conference in Chicago on Monday, Oct. 22, controversial comments from Jon Toigo, CEO and managing principal of Dunedin, Fla.-based Toigo Partners International LLC, had attendees shaking their heads in bewilderment.

For one, Toigo advised attendees against using VMware in favor of mainframe systems, calling the virtualization software “shoddy” in comparison. But nearly all conference attendees were assembled to learn about implementing virtualization in their data center.

Toigo is knowledgeable about disaster recovery (DR) planning, which was the topic of his keynote session, but he warned users against ever feeling completely secure with their disaster recovery plans.

More here.

As if there weren’t currently enough companies with a virtualization play, Hitachi has announced that it has placed it’s Virtage virtualization code on the firmware for it’s Xeon-based blade solution, the BladeSymphony 1000, which combines Itanium and Xeon CPU’s in a single chassis.

More here.

Nationwide Insurance is a posterchild for the right use of virtualization, on all their platforms. A huge IBM Mainframe shop, they also use Linux in their X86 environment to cut costs and be more green.

“While many IT shops see virtualization as a question of adopting EMC’s VMware on servers running Windows or Linux, Nationwide Insurance has adopted the technology for both x86-based and mainframe-hosted servers.  After all,” notes Buzz Woeckener, the company’s zLinux/Unix server manager, “virtualization was invented for mainframes.”

More on Nationwide’s virtual environment

I’d buy that for a dollar. Well, lots of dollars, since it’s priced similarly to our friends at IBM’s mainframes. PSI (Platform Technologies Inc.) makes a plug-compatible set of machines called the System64 DS and ES, formerly only available through a 3rd party.

DS or Distributed Server and the ES or Enterprise Server hardware are currently shipping to early adopters and are competitively priced with IBM’s per-MIPS pricing of approximately $1200 to $1400.

The key to PSI’s systems are the ability to not only run OS/390 (IBM’s Mainframe OS) but also Linux and Windows. The DS system sports up to 8 Itanium v2 CPU’s and the ES supports up to 64 Itanium cores. Both support the production versions of z/OS, OS/390, HP-UX, Linux including SLES and Windows Server versions.

“Our customers have repeatedly told us that standardization and virtualization are key to their business success and an imperative for their IT strategy – today and for the future. We’ve developed the System64 product line to meet the industry’s challenges and drive customers’ success.”

You can get a lot more information from the PSI site, including quotes and whitepapers.

Enjoy,

RossB

Linux, specifically SUSE Linux Enterprise, that’s why.

Techtarget has the first in what promises to be a series of articles reporting on the IBM SHARE conference, where a lot of new technology and software gets introduced and this year it was clear that Linux on the Mainframe (System Z) is growing the Mainframe market.

Read Robert Rosen’s first in the series of articles and keep checking for more, Mainframe Linux is the most exciting thing that has happened on that platform since, well, ever.

It could go without saying, but I’ll take the hit, SUSE Linux Enterprise Server is a runaway hit for the System Z Mainframes, we have over 85% market share and growing.

I do a lot of work with customers who have Linux on System Z, and IBM’s direction next year will focus more on server consolidation using the System Z as a platform for virtualization.  (Sounds of things ramping up in the background).

Enjoy,

RossB

Searchdatacenter.com has a great guide to understanding Virtualization in the Data Center online, it covers:

  • Management and Metrics
  • Saving Power
  • Disaster Recovery
  • Best Platforms for Virtualization
  • Server Virtualization Software
  • Support and Management Tools

This is just the right article to introduce someone to the concepts and tools involved in Data Center Virtualization.

Enjoy,

RossB

Here are some public success stories for customers using SUSE Linux Enterprise Server on the (zSeries) System z platform:

http://www.novell.com/success/america_first.html
http://www.novell.com/success/casas_bahia.html
http://www.novell.com/success/fnb_omaha.html
http://www.novell.com/success/ebs.html

Related blog entries of interest:

http://opseast.wordpress.com/2007/03/06/why-insurance-companies-like-linux/
http://opseast.wordpress.com/2007/02/28/mainframe-linux-vs-unix/

There’s also a whitepaper that talks about running SLES on System z:
Server Consolidation with SUSE Linux Enterprise Server on IBM System z

IBM released a new virtual Linux environment called “IBM System p Application Virtual Environment” or “System p AVE” into public beta today.  Neither way of referring to this new technology is particularly mellifluous or easy to pronounce, but regardless this is seriously good news for those who have System p hardware and want to take better advantage of Linux on those boxes.

System p already has a very healthy market share at 31.5% of revenue, according to the numbers for 2006 and System p AVE’s advent is sure to help grow this market share.  Now it’s possible to consolidate AIX and Linux on the same server, the supported systems and applications are listed briefly here.

Some of the notable capabilities of AVE are:

  • Ability to run most Linux X86 binaries unmodified on System p
  • Virtual X86 environment and file structure
  • Dynamic translation and mapping of X86 instructions to Power
  • Caching to improve system performance
  • No need to re-compile, translate or alter existing distributions

For more info on this beta, including the signup forms to get involved, visit the System p AVE landing page.

Powered by ScribeFire.

One of my running themes for presentations and customer visits is always the topic of Virtualization, ie: taking the resources of one system and running multiple instances of various operating systems in virtual machines on that system.  I am constantly looking for success stories and documentation that will help techies convince the bosses, bosses convince the board and for our team to keep up with advances on the topic in all areas, not just the customers we have here in the East Region.

I found a great IBM site that has the greatest concentration of System Z success stories I’ve seen so far, broken out into the industry areas and vertical markets.  If you can’t find the numbers you need to make a point on this site, contact us or leave a comment and we’ll be happy to help you evaluate and plan where you can use Virtualization on any platform.

Nationwide Insurance is a large company. They have over 158 Billion (that’s BILLION) in assets and rank at # 98 on the Fortune 100.   Their server environment was mostly distributed in nature and very costly to maintain.

Their distributed server environment had several issues that drove them to consider virtualization as a solution, including rising costs for cooling, power/electricity, expansion of datacenters, datacenter floor space. A study in 2005 found that 78% of the distributed servers were running at an average of 10% utilization, which left a tremendous amount of server capacity unused on a daily basis.

The typical method for dealing with needs for processing capacity was to over-provision with physical distributed servers, and the lack of flexibility made re-allocating resources to other workloads and especially to other physical locations extremely difficult.  Additionally, the time to deploy a new server measured in months, which made it hard to handle spikes in server utilization and needs for capacity required more guesswork than actual fact.

Part of the due diligence Nationwide conducted was a study to determine what the best virtualization scenario would be, and they found that using a pair of IBM Z900 mainframe systems, they could virtualize hundreds of distributed systems.  The initial plan was to virtualize 150 servers by the target date, and they ended up virtualizating over 250.  Nationwide moved ahead and performed the conversion of a large portion of their distributed systems onto the mainframes and the savings alone are projected to be in the tens of millions of dollars over the next three years.

An example of their newfound capacity and flexibility was the Super Bowl ad and the resulting traffic they got to Nationwide’s web hosting environment.  In the testing before the actual ad ran, the systems involved were increased dynamically to handle more than 22 times the expected traffic, all while testing other major applications at more than 5 times the promised Service Levels.

Significant cost savings were achieved in the following areas:
- Floorspace and Power costs dropped by 80%
- Headcount dropped by 50%
- Provisioning dropped from months to days
- Disaster Recovery times dropped from days to minutes
- Virtual Environment costs were roughly 1/2 of Distributed Environment costs

There are some excellent resources available for companies that want to investigate virtualizaton on the Mainframe, including the IBM Success Stories and References site.  Also of note is the ZJournal magazine and website, it features a number of interviews, reviews and reports about using System Z and Linux.  Articles of note are here and here.

Does your company have any mainframes?  Do you use Linux on them?  SUSE Linux Enterprise Server 9 and 10 work great on Mainframes, email us and we’ll be happy to answer any questions you might have!

Ross Brunson – rbrunson@novell.com

Follow

Get every new post delivered to your Inbox.