Who are you kidding?

A colleague pointed me at this. My colleague has responded here. In this politically correct world I won't post my real thoughts here - suffice it to say that the authors should read some of the links in this blog about skills, costs etc. Ignorance is no excuse.

by pwarmstrong September 29, 2005
Permalink | Comments (0) | TrackBack (0)

King Ferdinand & Queen Isabella would have asked Columbus for zLinux Skills

Ferdinand_isabella

I just returned from Barcelona and while many banks continue to run legacy work loads on the mainframe some of these banks in Spain aren't implementing Linux and Java mainframe applications as quickly as they might want to because they report limited Linux & Java on zSeries/z9 skills available in the local market to support these new work loads.

While IBM has begun an academic computing initiative at Universudad Politecnica de Madrid (UPM) and is working to create a special zSeries course one of the banks that I visited told me that they don't have the skills they need to support Linux and Java on the mainframe and that while they applaud the work at UPM it's insufficient to help them get the skills they need to run their shops.

The bank that I called on is a big supporter of the mainframe and wants to learn more about encryption of data on removable media and disk to better leverage the mainframe's enterprise security features but hasn't made the progress they want to moving Linux and Java work loads to this platform because they don't have the skills in house.

A place for customers to find people with new work load skills on the mainframe is the independent  user group SHARE who is teaming up with IBM to provide support for students and young profressionals interested in mainframe computing. SHARE & IBM are teaming up to provide support for students and young professionals interested in mainframe computing by creating a community called "zNextGen".

IBM is working with over 150 colleges and unversities worldwide which offer educational resources on the IBM mainframe and has pledged to deliver 20,000 zSeries trained professionals by 2010 including 10,000 students in China in the next 5 years.

Companies looking for mainframe skills - - especially in Linux and Java - - should contact IBM, SHARE or one of the 150 colleges and universities worldwide educating students on this platform.

by rjhoey September 22, 2005 in Systems Technology
Permalink | Comments (5) | TrackBack (0)

WinFS versus System/360

Interesting blog post from Guy Kewney, the i-Kew, over at vnunet.com.

He starts by talking to Microsoft WinFS and segues to some S/360 performance stuff:

I remember Hedley Voysey, doyen of computer commentators, back in the 1970s, once explaining that "in order to be taken seriously, a real-time interactive system must respond to a user query within two to three seconds." Any longer, and the user will naturally think it's broken.

Just to put that in context: he was writing about a system communicating at 1200 bits per second over a "glass teletype" terminal, to several hundred users, based on a System /360 mainframe from IBM with a total processing capacity very roughly equivalent to an original IBM PC AT and a set of 3350 disks and the important thing was, it was optimised for user response.

Its often good to look backwards in order to understand how the future will play out.

Even Bill Gates has moments of 360 envy. Read more about that in my tribute to Bob Evans.

by James Governor September 22, 2005 in History
Permalink | Comments (2) | TrackBack (0)

zOS Contest

Registration in our contest is now over 260 students from over 50 universities and colleges in the US and Canada.

To respond to some of the comments about my earlier post:
Why just z/OS.  Why not z/VM?

Why not advanced steps in lpars or mainframe accessories, CMS Pipelines or Rexx. A mainframe is so much more than just z/OS! Too bad the participants will not be able to see more than just the one OS.

The answer to all that is this is our first time running this type of contest.  We'll consider opening up to more topics in future iterations.

Is there anything on IBM's web site about signing up for this yet? When does the contest begin?
http://www.ibm.com/university/students/contests/mainframecontest.html

by Christopher Baran September 20, 2005
Permalink | Comments (25) | TrackBack (0)

Some Mainframe Links

Here are some mainframe links to enjoy

WebSphere and zOS suck, apparently. This from an ISV developer.

Alabama set to return to the fold, in order to cut costs? So much for this "mainframe migration".

holy terabytes batman! oracle behind 100 TB datawarehouse. that is big. seriously big.

When did you ever see "mainframe" and "retro chic" in the same headline?

by James Governor September 19, 2005
Permalink | Comments (2) | TrackBack (0)

IBM ships mainframe on time is not news

Is it just me or this press release headline kind of bizarre?

IBM System z9 Mainframe Shipping to Global Customers on Schedule

Yeah. Ok. If this was a Microsoft press release it would be news. "Microsoft Ships Product On Time".

But in the world of zSeries, meeting a schedule just isn't news. I have been covering the mainframe market for ten years now and the number of delays to product introductions can be counted on one hand. Maybe even one one finger.

Whatever next?

IBM System z9 Offers High Availability Features...

I recently credited the z9 launch team for its "PR mastery". I am not expecting this particular press release to have quite the same impact...

by James Governor September 19, 2005
Permalink | Comments (2) | TrackBack (0)

Big Ben Liked the Security of the z9

Big_ben

I just returned from a trip to England in which I presented to about 75 zSeries users sharing the z9 Product Announcement.

The customers were very focused on the enterprise security features of the z9. They really liked hearing about the IBM road map to deliver a z/OS based encryption facility later this year which will enable customers to encrypt their tapes without driving z/OS MIPs and software stacks meaning an economical way to ensure privacy, audit and compliance for publically identifiable information. They also liked the news of using ICSF to provide secure key management for the enterprise in case they need to access any of this encrypted information 5-10 or more years from now.

The customers talked about the fact that they are concerned about all of the replicated data sets and temporary data files they have flying around their networked & distributed environments and whether their current privacy, audit and compliance procedures are providing bullet proof, industrial strength, protection of personally identifiable information. There was discussion about the new laws pending in the EU in which any disclosures of personal information might have to be reported to the people impacted and the fact this could result in litigation, liabilities and more importantly damage to their brands. On top of this, some of these customers are concerned about the implications for data synchronization and backup & recovery of these replicated date sets and temporary data files. Many of these customers see the benefits of moving their applications back to the data (instead of the distributed model of moving data to the applications) for mission critical core workloads. Moving the data to the applications in a distributed environment still makes complete sense for the commodity transactions.

Bottom line, tortured data sets and the increased focus on automating privacy, audit and compliance play to the z9 and it seems the enterprise wide security roles of this platform are of more interest to some companies than the fact the z9 is about twice the size of the z990.

I enjoyed my stay in England as the customers were a pleasure to talk to but I sure was glad to get home to some American food.

by rjhoey September 14, 2005 in Systems Technology
Permalink | Comments (4) | TrackBack (0)

The Great Wall

Greatwallofchina_snow

In my last job as IBM VP of Sales responsible for Financial Services Sector accounts on Wall Street in New York City, I never gave much thought to that other Great Wall - - the one in China. Little did I know what to expect when I recently traveled to HuangShu, outside of Shanghai, China, to participate in the IBM System z9 Announcement. Although some of the banking customers that my team and I supported in New York were large as measured by the numbers of customers they served it was hard for me to get my head wrapped around the fact that a few of the state owned banks in China each had more customers than there are people living in the United States.

The customers at the announcement meeting represented some of the largest banks, manufacturers the public sector agencies. Not surprising, most of these customers were very focused on minimizing scheduled and unscheduled downtime because of the large number of customers these companies supported. These customers understood that the System z9 is more than a big server and that it's a holistic approach to system design. They liked the fact that this holistic approach is designed to bring together industry-leading technologies including: servers, operating systems & middleware, and storage and networking technologies. The appreciated the fact that the z9 can play a large role in the automation and control of the enterprise including security and systems management.

During this two day meeting, I was asked to do press interviews. While I assumed most questions would be about the new mainframe would be about its capacity, the reporters were more interested in the role of the System z9 plays in virtualization, open systems and collaboration in an OnDemand Environment. They wanted to better understand the virtualization capabilities of the z9 in optimizing resource use across a customer's I/T infrastructure, allocating enterprise-wide resources according to business priorities and the fact the z9 continues to build upon the high levels of resiliency offered by the z990 predecessor to the z9. Discussions ranged from business resiliency delivered via parallel sysplex, GDPS for z/OS, GDPS Multi Sysem support of z/VM supporting Linux and HyperSwap, intelligent workload management and business integration using IFLs and zAAPs for Linux and Java integration to back end z/OS database access.

What surprised me most was the fact that some of the companies in China don't include people costs in their total cost of computing analyses because the price of labor is low relative to the costs of hardware and software which is the opposite of what we see in other more mature economies. While it was explained by some of the customers that because of low labor rates  in China that people costs are sometimes not considered in platform selection decision and total cost of computing analysis - - which could put the z9 (for which one can grow MIPs substantially with limited staff increases due to the advanced automation, systems management and virtualization of this centralized solution) at a disadvantage when compared to managing racks of less expensive Unix and .Net servers which can require large staffs to manage - - these same customers explained that when designing banking solutions to support 100 million or more customers the z9 is likely the only platform they will consider going forward because of it's high availability and quality of service.

These customers know that when their computer systems go down the lights go out in their businesses and as such they are willing to pay for the differentiated value this platform creates. It was clear these customers think that trying to support a bank consisting of 100 million accounts with racks of Unix computer is like trying to plow a field with hundreds of chickens harnessed together instead of a using a single ox; not a very smart approach and certainly one in which it's hard to get the desired results.

Other customers from smaller firms told me they too have mission critical applications so while not interested in our largest z9 models they are interested in smaller z9 models because they need the quality of service the z9 offers. The bottom line is that the customers were all telling me that if their computer systems go down and their companys' reputations for service is tarnished then they could lose their jobs. This isn't any more complex then the fact that these I/T profressionals need to buy a computer that works; all the time, because they need their jobs. They don't really care how the z9 computer works provided it works just dandy.

These customers reacted positively to IBM's continued investment in skills for the IBM Mainframe Community and IBM's advancing toward the goal of helping train an additional 20,000 skilled mainframe professionals in the marketplace by 2010. Some of these customers were aware that IBM recently built a high-availability center, as part of our Academic Initiative for z9, and that we shipped five mainframe systems which are shared by seven Chinese universities.

I left HuangShu heading back to Beijing with the belief that the customers at this announcement understood that the IBM System z9 wasn't just about the size or speed of the machine but that the enterprise role this machine serves in providing the next level in computing scalability, improved computing economics and security.

China is an amazing country and as it continues to grow as a global economic power I think that we'll see a lot of z9 machines being purchased to support this new economy.

by rjhoey September 14, 2005 in Systems Technology
Permalink | Comments (1) | TrackBack (0)

Enterprise Data Centre Newsletter

Just got this:
"Arcati has re-launched its Enterprise Data Centre newsletter, and the September issue can be accessed at http://www.arcati.com/EDCEsept05.pdf. Our main target readership is mainframe-focused data centres in Europe but, as you can see, much of the content is relevant to large systems users worldwide."

by pwarmstrong September 7, 2005
Permalink | Comments (1) | TrackBack (0)

SOA and Mainframe

I have just read this:

Mainframes 2005: Politically Correct and Good for Business

At last week's SHARE User Events in Boston, the editors of 5 Minute Briefing had the opportunity to meet and speak with industry practitioners and vendors from all points of the data center. We've come away convinced that the mainframe has seen a renaissance, and is now the platform of choice for open systems and open standards (SOA) initiatives. As one IBM executive pointed out to us, the word "mainframe" may have not been politically correct a few years ago, but more and more vendors now are trying to point out how "mainframe-like" their products are. For more insight into issues that affect data centers, subscribe to Database Trends and Applications magazine at www.dbta.com/subscribe.html.

SOA - agree, bring it on.

Mainframe-like - anyone like to supply a definition? Here are a few bits from Wikipedia:

Modern mainframe computers have abilities not so much defined by their performance capabilities as by their high-quality internal engineering and resulting proven reliability, "expensive" but high-quality technical support, top-notch security, and strict backward compatibility for older software. These machines can and do run successfully for years without interruption, with repairs taking place whilst they continue to run. Modern mainframe computers have abilities not so much defined by their performance capabilities as by their high-quality internal engineering and resulting proven reliability, "expensive" but high-quality technical support, top-notch security, and strict backward compatibility for older software. These machines can and do run successfully for years without interruption, with repairs taking place whilst they continue to run.

...corporations found new uses for their mainframes, since they can offer web server performance similar to that of hundreds of smaller machines, but with much lower power and administration costs...

Financial institutions have not experienced these security-related failures with their mainframes, so many organizations are reassessing their entire data handling practices, often focusing on data recentralization on secure systems

etc etc etc

by pwarmstrong September 7, 2005 in Future
Permalink | Comments (0) | TrackBack (0)

On CICS blogs and blue satin sheets

This post comes to you a few days late. I had an awful experience using Typepad as a text editor (DOH!) last Friday, when a single mistaken mouse click meant I lost about an hour's work. I won't even try and repeat the work. Suffice to say I wanted to do some linking, to encourage community clusters. That is, if you link to someone they are more likely to link to you and yours.

I just came across this rather nifty blog on CICS by IBM's Ian Mitchell. I would recommend you subscribe if you're interested in mainframe for SOA.

Its also good to see a skeptical post about the value of the z9 investment. Note to readers: it got 17 comments. That is more than any post on this mainframe blog so far. May i suggest you take the time to comment, to call us on BS, to add context where its forgotten, to tell us something new. The comment system is very easy to use, and we would love to hear from our readers. If you have an opinion why not get involved. A blog is only as useful as its community.

Talking of communitiy what about the zEcosystem? CA wants to go toe to toe with IBM in terms of zSeries feature development, according to Sam Greenblatt. New CA CEO John Swainson (ex-IBM) knows just how valuable the mainframe franchise is, but more importantly how important innovation is to sustain that value over time.

I like the way John Patrick looks backward to look forward, calling out similarities between IBM in the "bad old days" and Microsoft. He says, in context of Massachussets decision to drop Microsoft file formats:

We used to say that with mainframes there was never a problem exchanging information between computers -- as long as each computer was a mainframe!

Sometimes the parallels between IBM then and Microsoft now are scary.  Everything talks together as long as your end points are single vendor solutions based on "integrated innovation"...

I hope you don't mind me also pointing to a one of my blogs from last week that hit a mainframe theme. I argue for the importance of software currency:

Currency is a very clear indicator of ongoing commitment from an installed base. In many respects currency is far more valuable as a market barometer than installed base because it allows us to consider potential market inflexion points. Do any of the analyst market share counting companies parse currency in their methodology? Not that I know of. RedMonk certainly pays close attention to the factor.

Currency is one reason the folks at IBM are happy about about CICS futures at the moment. zSeries customers are swiftly upgrading to the latest versions of the transaction management software, because they want to move forward with service oriented architectures. The flipside - those mainframe customers that are not getting current, that aren't modernising their infrastructures, are the ones most likely to fall to competitive migration programs by other vendors.

Looks like T-Systems South Africa is the latest organization to choose zSeries as a Linux host.

Here is a really nice take on open source. For all those newbs that think open source is somehow new - you need a history class. At IBM all software was open source until 1969.

And finally: did you know the new System z9 looks just like a blue satin bed sheet?

by James Governor September 6, 2005
Permalink | Comments (2) | TrackBack (0)

Just Enough Is Good Enough

We used to say that with mainframes there was never a problem exchanging information between computers -- as long as each computer was a mainframe! The point was that IBM mainframe computers really excelled at compatibility. A customer could order a complete mainframe system for a new location. The central processor would arrive from Poughkeepsie, the tape drives from Boulder, the disk drives from San Jose, the card reader from Endicott, the network controller from Raleigh, and other components from IBM factories all over the world. After the IBM CE (customer engineer) got everything hooked up and tested, it would be turned over to the customer. Files could immediately be exchanged with other company locations. Everything worked because IBM assured 100% compatibility by controlling all the interfaces. This was good. Microsoft is trying to do the same thing with Windows. This is not good. What's the difference? The Internet.

The Internet has changed everything. It is built on open standards and increasingly systems and procedures and data which use the Internet are alos becoming open. I am quite excited about what OASIS is doing with the OpenDocument standard. Massachusetts may be just the first of governments around the world who are mandating OpenDocument. From my perspective it is a matter of "Just Enough Is Good Enough".

by John R Patrick September 3, 2005 in History, Systems Technology
Permalink | Comments (2) | TrackBack (0)



The postings on this site are our own and don’t necessarily represent the positions, strategies or opinions of our employers.
© Copyright 2005 the respective authors of the Mainframe Weblog.