Published: January 31 2010, 09:36 AM | 1 Comment(s)
by Reg Harbeck

I remember, 15 or more years ago, seeing a photocopy in one of my colleagues' cubicles, with a picture of a dinosaur running rampant eating little PC-like creatures.

A few years later, I saw a similar graphic on some mainframe services buttons that were being given out.

The message on both? Mainframes: We're Back and We're... (insert word suggesting "in a mean mood").

In other words, we mainframers have known it for years.

The good news? The rest of the world is finally getting it too!

One of the most recent indications of this is the following article in the Economist that came out earlier this month: http://www.economist.com/businessfinance/displaystory.cfm?story_id=15276714

Now, more good news: the Arcati Mainframe Yearbook for 2010 is out, and it's packed with the kind of stuff that warms a mainframer's heart (dinosaurs? maybe; cold-blooded? definintely not!).

You can see the yearbook website at http://www.arcati.com/newyearbook10, or just go straight for the PDF at http://www.arcati.com/yearbook10all.pdf.

You know, it just keeps getting better and better to be a mainframer!

Share this post:  Email

Share

Published: January 29 2010, 09:44 AM | no comments
by Marcel den Hartog

Many moons ago, most Mainframe applications went through an audit to see if they could be re-hosted on a platform that was cheaper, more flexible and more powerful.

It was a business decision because (on paper) the cost of running the migrated systems was looking great. Most IT people know what happened. We managed to "re-host" the applications that could be re-hosted, but at what cost?

Everybody in the industry knows that the projects were long and the results mixed. Over Promised and Under Delivered. And the REAL cost? The real cost of hardware, managing, life cycle management, security, disaster recovery, the cost of expanding and changing the application? Some projects I am aware off were 5 times as expensive and cost many managers their jobs. And these were the applications we decided would be the best candidates to re-write!

Now don't get me wrong. If it makes business sense to rewrite/re-host an application, by all means do it. But do it for a reason, and not because you CAN or because someone tells you it's cheaper!

A recent series of "Modernizing Mainframe environments" articles states: "Others have decided to migrate off of the mainframe entirely, opting for immediate savings with Linux, Unix or Windows environments." Cool! Immediate savings! I wonder how THAT works... No migration of data? No rewriting of code? No new functionality in the process? No! Heck, why would you want to do that?? The Mainframe is old, so the application running on it is old. Just migrate and modernize and achieve immediate savings!!

It sounds like two characters on TV telling each other:

"Bob, what is that strange screen?".
"Bill that is a 3270 Application..."
"But Bob, that looks old fashioned and out of date! Just green letters and numbers!"
"Bob, I have a small device that can transform your application into a sexy looking application, would you want that, Bob?"
"But Bill, how does that work? Is that not expensive?"
"No Bob, it's free, and what is even better, you will have immediate savings!!!!"

If you believe that, you will also believe that you can have a six-pack in three weeks by training just 2 minutes a day.... Overpromising, remember?

Our industry is famous for getting approvals for projects like this based on low estimates and ending up with something that is five times as expensive and has 60% of the functionality. We are FAMOUS for that! The applications that still run on the mainframe today are the most complex pieces of work you have ever seen. They do what they are supposed to do, have years (sometimes decades) of knowledge in them, perform like a jet-fighter and are among the most reliable assets your company has. Treat them with respect... Modernizing can often be done without rewriting everything.

If you want to re-host systems (from Mainframe to Distributed OR from Distributed Systems back to the Mainframe), don't listen to stories like the one I just described. Don't accept a plan without a full architecture that includes Management, Disaster Recovery and Life Cycle Management. The times this could be added as an afterthought are behind us. Ask for a real ROI, and one that has been looked at by people from different disciplines, not something thrown together by 1 or 2 people... And expect it to cost 3 times more than estimated... Be real...

Share this post:  Email

Share

Published: January 15 2010, 03:18 PM | no comments
by Marcel den Hartog

Life of a "user" (you know, the people who actually use our IT Infrastructure to do their day-to-day work) is not simple. Not only does our company expect them to do the work they are assigned to do, but we often do not realize how tough their life really is. I spoke with a large customer yesterday and he summed up the IT Systems and Applications they had running. More than 18.000 Servers (every possible flavor), and two Mainframes running a mix of legacy and "modern" applications. Almost all with different interfaces and often not as easy to work with as we like to think..... Some users in this particular company had to work with more than 15 different applications, all with different interfaces. It's workable, people live cope with it, until there is a performance issue......

The first question often is: is there really a performance issue? Some applications work faster than others and sometimes it's the perception of the user. So the user spends some time figuring out if it's him/her, the PC he/she is working with, or the application. After he/she concludes that it must be the application, he/she opens an issue in the Helpdesk. And then they start receiving emails about changes and progress. The Helpdesk investigates, and if they have the right tools, they can sometimes confirm that there IS an issue. Often, this requires 1-2 calls with the end-user ("try this", reboot, "try that", etc.). But then, the real road to hell starts.... Without the right tools, many departments will be involved to investigate: administrators, operations, network experts, application experts, Distributed database experts, Mainframe operational experts, CICS experts, MQ specialists, DB2 DBA's and many, many more. And every time, the end-user will receive status changes in his mailbox and often will be asked to perform little tests.

In the meantime, more and more users will experience the same problems, call the helpdesk, open new tickets with slightly different descriptions etc. Mayhem... And we wonder why end-users sometimes lose confidence in our IT Infrastructure....

The fact that we are firefighting in our little silos is often very obvious to our users. And it costs our company a lot of money.... We all know that the first step to fight firefighting (....) is automation. Use solutions that monitor trends, solutions that are aware of what the baselines are and that warn us when things start to change slightly. Because this is often a sign that soon, things will start to go wrong. And these solutions should be smart enough to tell the Helpdesk staff AND our Level 1-2-3 staff where the problem occurs. Not what component is experiencing problems, but what application is having issues. So we speak the same language as the user.... This way, we can investigate what to be investigated, and not what we think should be investigated... Today's applications cross multiple networks, platforms, databases and protocols. Manual monitoring simply is not up to the task and only an integrated set of Application performance Management solutions that uses automation to monitor and alert can help us to keep the lives of our end-users (our customers) simple. Find out more about solutions that will not only make your life easier, but that will help you to become a hero in the eyes of the end-users:

http://www.ca.com/us/webcasts/ondemand/item.aspx?e=218105&eis=1

Share this post:  Email

Share

Published: January 07 2010, 09:41 AM | no comments
by Reg Harbeck

Once upon a time there was a rainbow. Not just any rainbow, mind you: this one had colors ranging from Aqua to Tan to Silver. But to many, the most important color in this rainbow was also the second color in the traditional ROY G BIV rainbow: Orange.

In case you haven't guessed yet, I'm referring to the Rainbow Series books (sometimes known as the Rainbow Books). They were a series of computer security standards published by the United States government in the 1980s and 1990s. According to Wikipedia (see http://en.wikipedia.org/wiki/Rainbow_Series), they were originally published by the U.S. Department of Defense Computer Security Center, and then by the National Computer Security Center.

The Orange Book, so named for the color of its cover, was of particular interest to those who provided and used Identity and Access Management software such as CA ACF2 and CA Top Secret. Its official title was DoD Trusted Computer System Evaluation Criteria (aka TCSEC; see http://en.wikipedia.org/wiki/Trusted_Computer_System_Evaluation_Criteria) and it spelled out what criteria were necessary for a security product to meet various levels of security, signified by a letter (for "divisions") and sometimes a number (for "classes"). At the bottom was "D" for "Minimal protection." Then, C1 and C2, two levels of "Discretionary protection." Next, B1 through B3 for various degrees of "Mandatory protection" and at the top, A1 and beyond for "Verified protection."

CA ACF2 and CA Top Secret were evaluated and certified for configurations that were C2- and B1-rated.

Then, everything got superseded as TCSEC was replaced by the Common Criteria international standard originally published in 2005 (see http://en.wikipedia.org/wiki/Common_Criteria).

However, CA is nothing if not customer-centric, and we have some very important customers who subscribe to these criteria and wanted the assurance that we continue to be compliant.

So, as we announced today (see http://www.ca.com/us/press/release.aspx?cid=225622), that's just what we've been doing for CA ACF2^TM r14, CA Top Secret® r14 and CA Compliance Manager for z/OS r1.

What does this mean to you? Likely, not a thing, because it's the same software you've been using all along. We're just getting certification (again) that it really is as good as we say it is.

But, just in case you care about this kind of thing, or, like me, like to keep your stock of trivia topped-up, now you know the color of CA's mainframe security products. Orange you glad you asked?

Share this post:  Email

Share

Published: January 05 2010, 03:11 PM | no comments
by Marcel den Hartog

Mainframers,

In the past months I witnessed a phenomenon. Cloud is hot, and the hype-cycle is running at full speed.

And some of our fellow mainframers see the opportunities: "The Mainframe In The Cloud", "The Cloud & The Mainframe", "10 Reasons Why a Mainframe Is A Perfect Cloud". And all the articles state the obvious: The current IBM z10 Mainframe is a perfect candidate for what is called "Cloud Computing".

Good!! Fantastic!! But who exactly is it we are explaining this to? I get the impression that we are mainly talking to each other about the virtues of the mainframe. We are talking to believers. Which is easy, because WE all agree.

So let us first agree that we all agree. The Mainframe is the best platform to host one or more clouds!

Take some time to recognize the momentum. Everybody wants "cloud". And at the time we are telling other mainframers how good the mainframe is, the "Blade people" are building clouds. They are demonstrating to Architects and IT Management how easy it is to run 12 images on 2 or 3 physical boxes and they demonstrate it as a mini-cloud. More a puff of smoke, but the architects and your IT Management sees "Cloud" in action. And they like what they see. So the architects start making plans to build the next generation IT Infrastructure. And in these plans, I do not see the "Z" word... Believe me, I see it happen around me..........

I have learned an important lesson in the past: If you want to convince people about something, you start by talking about it, but you very quickly switch to actually DEMONSTRATING the point you are trying to get across. So let's stop talking. Fire up some Linux and demonstrate that you can run the same stuff as "they" can. More flexible, more efficient, more powerful and better managed. Demonstrate that you CAN get 20 year old virtualization techniques to work for your company. Demonstrate value... The tools to manage the virtualized environment are available, so that should not stop you (http://www.ca.com/us/mainframe-linux.aspx).

I know that this is not easy. I realize that we are doing more with 8 people than the "others" do with 200. I know your bandwidth is limited and there is a machine that needs to keep running in order to keep your company going. But if we now fail to demonstrate (in a pragmatic way) what we CAN do, this opportunity might pass and who knows when the next one will come around. IF it ever comes around. ...

Share this post:  Email

Share