Triton Consulting

Latest Webcast from IBM

This session will cover the end-to-end monitoring, alerting and protection of information that’s stored on or managed by System z. You’ll see how it can help:

• Ensure compliance by properly controlling who sees, uses and accesses your data
• Monitor usage and access and detect threats in real time
• Protect the underlying DB2® and IMS™ data stores
• Ensure separation of controls and flexible audit and reporting
• Archive data intelligently for compliance, performance optimization and lower storage costs

In his column for Database Journal Julian Stuhler looks at how an upgrade to DB2 10 can be justified.

Under current economic conditions IT professionals are being asked to give short-term business benefit cases before any major infrastructure upgrade is agreed.

DB2 10 has a lot of interesting new functions but the more immediate benefit is the promise of significant CPU reductions for most OLTP, query and batch workloads, with initial testing showing savings of 5-10% out of the box with no application changes, when compared to Version 9.  As most DB2 for z/OS csutomers are on usage-based charging, CPU reduction tranlates directly into lower monthly fees for the DB2 licence – an immediate business benefit. 

Click here to read the article in full.

As always, the answer to this is “it depends”! However, here are some basic rules of thumb that you may find useful:

• Define an application environment specifically for your Java stored procedures. This should have NUMTCB set to a relatively low value (6-8 is good) as Java stored procedures typically require a lot of storage. Java SPs also have their own JCL DD statement requirements, such as entries for JAVAENV and JSPDEBUG.
• Define another environment specifically for your REXX stored procedures. Only a single REXX SP can run in any given application environment at any one time, so you need to set NUMTCB=1 for these.
• Define an application environment specifically for stored procedures that need to run under RACF program control (some of the IBM-supplied SPs fall into this category, such as DSNACCUC). These SPs need their own environment: if you try to execute both program-controlled and non program-controlled SPs in the same environment you’ll receive message ICH420I and the controlled SPs will fail.
• Define an application environment to host all of the IBM-supplied stored procedures that don’t fall into one of the above categories.
• Define 1-n environments for your “normal” COBOL, PL/1, C or SQL stored procedures. NUMTCB for these can typically be set fairly high (40-50 isn’t uncommon).
• Finally, you may want to consider having one or more test environments for your normal procedures, but with NUMTCB=1. This forces procedures to be run individually, which can be very useful during debugging when your procedure writes messages to address space log. Failing to do this can result in several procedures interleaving their debug messages, which can get very confusing. A simple ALTER PROCEDURE statement is all that is required to change the application environment, but don’t forget to stop/start the procedure afterwards.

Webcast from IBM.

April 20, 2010, 11 a.m., Eastern Daylight Time, 3:00 p.m. UTC

The life of a database administrator (DBA) is not easy these days. Database performance issues, worries about data corruption and loss, expanding business requirements, and rising chargeback costs are just a few of the unnerving trends that are causing DBA personnel to be stretched further and further — and closer to the breaking point.

We invite you to attend this complimentary teleconference to learn about the best practices and success stories of companies in a wide variety of industries who have turned their DBA staffing around with the help of IBM DB2® solutions on z/OS®. Over-burdened DBAs can get help and valuable support by using DB2 solutions for performance management, administration, change management, backup and recovery, cloning, and data governance. As a DBA, it puts you in a position to lower operational costs, increase your productivity and help your enterprise prepare for its future success.

In this teleconference, you’ll learn about DB2 solutions that can help:

• Monitor, analyze and tune the performance of DB2 and DB2 applications on z/OS
• Easily navigate the DB2 catalog, manage and track changes, and propagate them
• Perform automated backup and recovery operations and eliminate errors in manual processing
• Automate the cloning process and have usable DB2 clones within minutes

I’d like to follow up on a topic last mentioned in this very column last November, on the subject of “Not Accounted” time (link here ). The list of possible explanations for not accounted time that Willie referred to was very comprehensive, and included some scenarios I’d never come across before. Recently however, I was involved in a situation that wasn’t covered in that or any other material I’ve ever seen on not accounted time, and it had us scratching our heads for a while….

Visit the blog

March 2010- Unintended Consequences

We’ve all done it.  You see a great new feature in DB2 and jump headling into using it without fully appreciating all of the implications.  Here’s a case in point:

One of our clients is a large SAP customer running on DB2 for z/OS V9. With a huge number of subsystems to support and over 35,000 tables in each, taking traditional image copies would be a bit of a nightmare. Luckily, DB2 9 for z/OS contains the BACKUP/RESTORE SYSTEM function, which uses Flashcopy technology to rapidly backup and restore an entire system with a single utility. That means no messing with the traditional tablespace-level COPY utility, except when it’s needed for other reasons (such as when taking an inline image copy during an online REORG, to avoid putting the tablespace into COPY PENDING). 

So far, so good. However, a couple of weeks ago the DBAs came in one morning to find a lot of SAP users complaining about database errors, and dozens of tablespaces unable to be updated due to being in a COPY PENDING state. What the heck happened?

Visit our DB2 Geeks to find out

If you make use of RTS and DEFINE NO, you might want to review your processes to ensure that you’re picking up all of the tablespaces you should be……

Since V7 of DB2 for z/OS, IBM has provided a Real Time Statistics (RTS) feature, which is a set of tables that DB2 keeps (more or less) up to date with recent activity against all tables and indexes in the system. In particular, the RTS tables provide detailed information on the amount of update activity that has happened since a given object was last REORGed or had RUNSTATS executed against it. That information is obviously very useful when determining when those utilities need to be run next, so many sites base their housekeeping schedules on this information (either using their own bespoke code and procedures, or a vendor tool such as IBM’s DB2 Automation Tool). In particular, looking at the number of INSERT/UPDATE/DELETE operations as a proportion of the total rows in the table (using the TOTALROWS column) is useful, so you can reorg a table when more than 10% of the rows have been changed, for example.

We came across a problem recently, where one of our customers was using this feature in a DB2 9 for z/OS subsystem that had lots of DEFINE NO tablespaces (which instructs DB2 to defer creating the underlying VSAM dataset until the table is actually referenced – very handy for ERP systems such as SAP that create a large number of tables that are never used in a given environment).

For normal DEFINE YES tablespaces, a row is created in the RTS tables as soon as the tablespace is created, with the statistics initially set to zero values and updated in real time to reflect activity against the underlying table. Unfortunately, the situation is different for DEFINE NO tablespaces: the row isn’t created until the VSAM dataset is defined on first use (no problem there) but instead of zeroes some fields (including the number of rows in the TOTALROWS column) are set to null instead. That null value isn’t updated until the first REORG or LOAD is run against the table, so if the table is populated via INSERT rather than LOAD, your nice RTS monitoring routines may never trigger that first REORG!

This “feature” has been reported to IBM, but in the meantime if you make use of RTS and DEFINE NO, you might want to review your processes to ensure that you’re picking up all of the tablespaces you should be……

The 2010 Arcati mainframe yearbook is now available to download here.

This year’s topics include:

• Mainframe continuity planning
• Leveraging existing mainframe investmentsin modernisation projects
• Lowering mainframe TCO through zIIP speciality engine exploitation
• SNA mainframe security
• TurboHercules and disaster recovery – an innovative approach to mainframe outage and business continuity
• Software migrations

Managing performance and cost has become a significantly more difficult job with capacity planners and performance analysts being asked to defer hardware and software upgrades due to squeezed budgets. Many large IT departments have also seen loss of staff due to redundancy and this has put added pressure on maintaining good application performance.

One of the most effective ways of reducing or containing mainframe costs is through better managing CPU consumption. By slowing down the growth of CPU usage, hardware and software upgrades can be deferred while often improving performance thereby allowing organisations to keep costs down and performance and profitability up.

MIPS Growth in the Mainframe Market
In a recent mainframe market study, IBM reported that the mainframe has seen a 20% compound annual growth rate in MIPS since 2003. Also, In Ovum’s “The state of the mainframe” research, they found that mainframe MIPS growth is averaging around 20 percent per year and large mainframe-centric enterprises have been consistently averaging 35% plus MIPS growth.

Whilst it is good news for businesses to see transactions on the rise, with usage based pricing for z/OS this increase in workload pushes up software costs and can also negatively impact application performance.

The effects of growing MIPS
Performance – Typically, any significant increase in the amount of CPU used by a given workload will result in an associated increase in transaction elapsed times. For performance-critical online workloads, that increase can translate directly into poorer critical business metrics such as customer satisfaction and retention.

Just throwing more MIPS at a poorly-performing DB2 workload does not always address the issue. A 2 hour response time may be reduced to 1.5 hours with more CPU time being available, but the problem might be due to a poor access path and some DBA attention could get it down to 5 seconds. This is especially true of application performance tuning, which is where the majority of performance issues tend to lie.

Cost – Although performance is a key issue for many organisations, the major diver for many IT teams is the need to reduce mainframe resource usage and thereby potentially defer hardware upgrades and reduce monthly MIPS costs. There is also human costs to consider: maintaining an underperforming system takes more time and resource for IT teams and adds pressure from the business teams who are calling for improved response times.

Can tuning do enough to reduce costs?
The majority of customers have significant potential for reducing resource consumption through tuning. This is especially true for those with older applications that haven’t been actively maintained for a while or who have lost some of their deep DB2 skills through retirement or redundancy.

By implementing key tuning procedures, ongoing software costs can be reduced and mainframe upgrades deferred. In addition, application performance will be enhanced and overall TCO reduced.

Check out the full article on IT-Director.com or download the white paper

Are you moving to Version 9 in 2010?  Attend the DB2 Symposium on 8-9th March for the DB2 for z/OS Version 9 Transition Class in London.

Look out for your 10% discount code in our newsletter coming out shortly.  Don’t currently receive our newsletter?  Subscribe here  . 

Already subscribe to our newsletter but can’t wait to book?  Email us today and we’ll send you your discount code.