1. Cheryl Watson’s Tuning Letter 2011 No. 5
2. Correction to Tuning Letter 2011 No. 5
3. RSM Problem?
4. Two IBM Red Alerts
1. Cheryl Watson’s Tunig Letter 2011 No. 5
The 41-page 2011 No. 5 Tuning Letter was emailed to paid subscribers on December 20, 2011. You may visit our website at www.watsonwalker.com to obtain subscription information and the table of contents. The following is the Management Summary from that issue, talking about some of the contents of this latest Tuning Letter:
The majority of this issue relates to new functions that are identified as being part of z/OS 1.13. z/OS 1.13 became available on September 30, 2011, and contains many significant improvements, especially for large installations. But even small installations can benefit. However, most sites are at least six months to a year away from installing this release. So why do we spend so much time and space on it?
It’s because many of these new facilities are available in the normal maintenance stream for z/OS 1.11 and 1.12, and are ready to exploit today. So our Focus article,z/OS 1.13 Goodies Now, includes all of the new facilities that you can start using to-day. A few might need additional hardware, such as zHPF (High Performance FICON) or DS8000 storage server enhancements, but the savings often justify the cost of the new hardware. Wouldn’t you like to reduce your batch window significantly, or reduce the response times of your online systems? Another important part of this Focus article is the summarization of IBM statements of direction (SODs). For example, z/OS 1.13 is the last release to support a staged migration of z/OS and JES. A staged migration refers to the fact that currently, you can install z/OS 1.13 and continue to run JES 1.12, but in future migrations, you will need to migrate (for instance) to z/OS 1.14 and JES 1.14 at the same time.
A second major article contains all of the changes to parmlib members that were made in z/OS 1.13 documentation, but actually apply to earlier releases (from z/OS 1.10 through 1.12), especially if you applied maintenance regularly. There are several reasons why we provide this, but it’s mainly to provide recommendations on which new features to exploit and which to avoid. We also do this work because IBM identifies only a few of the actual changes. We had to compare several manuals to identify the actual changes. This section should reduce the work for your sysprog or performance analyst significantly.
Windows on zEnterprise
When IBM announced the zEnterprise Blade Center (zBX) for consolidating AIX serv-ers, customers immediately started asking for the ability to consolidate Windows and UNIX servers. So on April 12, 2011, IBM announced (111-078) they would support running Linux and Microsoft Windows on System x in the zBX. Support for Linux was announced (111-129) on July 12, 2011, and support for Windows was announced (111-167) on October 12, 2011. The Windows support is provided on IBM BladeCenter HX5 (7873) blades, and is available only when the z/OS system is on a z114 or z196 machine. For more information, see the press release on November 7, 2011 – For the First Time, IBM Brings Mainframe and Windows Together – www.ibm.com/press/us/en/pressrelease/35925.wss. With the huge number of Windows and UNIX servers, the ability to consolidate these and manage them as a single group provides improved security, reliability, and performance. And at the same time this reduces the staffing requirements. IBM claims that they have reduced the total cost of ownership (TCO) in their own centers by 70% using the zEnterprise consolidation.
Elsewhere in This Issue
You’ll also find many other useful items throughout this newsletter: When and why to use a zBX • Caution about using a DB2 APAR to improve performance • Significant enhancement for zEnterprise High Performance FICON (zHPF) • Important publications and papers from IBM • SMF and Information APARs to help you identify useful maintenance.
2. Correction to Tuning Letter 2011 No. 5
Jack Schudel of the University of Florida pointed out that, on page 37, I attributed two SHARE presentations to John Hutchinson instead of the actual presenter,Gregory Hutchison. My apologies to both. The information below is correct.
IBM often adds presentations from SHARE and other conferences to the ATS website:
TC000094 – SHARE Orlando Session 9025 – zBX Hardware Overview. Gregory Hutchison (19Nov2011)
TC000095 – SHARE Orlando Session 9036 – Server Time Protocol Overview and Planning Considerations. Gregory Hutchison (19Nov2011)
3. RSM Problems
A customer reported an interesting problem, but IBM has not opened an APAR on the problem. So we’d like to ask you to take a look and see if this situation exists in your site.
The problem is identified in RMF Paging Reports where the Hiperspace central storage frame counts continue to grow after an IPL. It’s even possible for these values to exceed the actual number of central storage frames on the system. It’s also noticeable because the MIN, MAX, and AVG frames are very similar continually. The solution of the PMR is to recycle the initiators that run large DFSORTs that use Hiperspaces. A working hypothesis is that an RSM control block (IARRAX) field, RAXHRECT, is not getting decremented under some conditions (possible when a standard hiperspace page from a large DFSORT is moved from a real frame to an auxiliary storage slot).
The problem is still being investigated by IBM, but we’d like to know if anybody else is seeing these large values in their RMF (or CMF) reports. Please let us know at firstname.lastname@example.org. Thanks.
4. Two IBM Red Alerts
IBM issues Red Alerts for extremely critical problems. You should consider subscribing to their service at www14.software.ibm.com/webapp/set2/sas/f/redAlerts/home.html. Here are the latest two (and there were only three others this year):
2011.11.16 – Fix for APAR PM51655 is needed on DB2 10 for z/OS data sharing environments as a prevention of potential data loss. PM51655 (DB2 10, HIPER, 30Nov2011) – Various Data Inconsistencies With Data Sharing at V10 Since Some GBP Dependent Pages Were Not Castout.
2011.10.28 – Fix for APAR OA37264 needed on z/OS 1.10 and higher for exploitation of greater than 64 subchannels to a Coupling Facility when running on z196 GA2 or z114 at Driver 93G. OA37264 (z/OS 1.10-1.13, HIPER, 16Nov2011) – ABENDS026 RSN0C1C1026 and ABENDS026 RSN09340006 After a Dynamic Add of a Coupling Facility (CF) with More than 64 Subchannels. The situation described in the Red Alert and APAR could result in a system or sysplex outage.