1. About Cheryl Watson’s Tuning Letter 2011 No. 1
2. Correction to Tuning Letter 2011 No. 1
3. SHARE in Anaheim
4. SMF Type 113 Records
1. About Cheryl Watson’s Tuning Letter 2011 No. 1
The forty-five page 2011 No. 1 Tuning Letter was emailed to paid subscribers on February 24th. You may visit our website at www.watsonwalker.com to obtain subscription information and the table of contents. The following is the first page from that issue, talking about some of the contents of this latest Tuning Letter:
This issue starts us on our 21st year of Cheryl Watson’s Tuning Letter! Thank you all for a wonderful twenty years. We couldn’t have done it without your feedback and suggestions. I thought I’d be retired by now (I’m 66), but this is so much fun, I can’t see retiring any time soon. Some of you have been with us for all twenty years, and we hope you’ve enjoyed learning and contributing as much as we’ve enjoyed sharing.
And thank you so much for the feedback on our last issue where we started this series on parmlib changes. Most agree that they simply haven’t taken the time to look at exploiting new options in parmlib. Others have said that they found some very poor settings in parmlib members that haven’t changed in a decade. It will still take another issue to complete all of the parmlib members, but it’s been fun doing this review. Be sure to tell us of any of your recom-mendations or experiences.
The 2010 DVD was mailed to subscribers on February 9th. And here’s a new hint – at the end of a search for a topic, change the sort order from ‘Relevance Ranking’ to ‘Filename’. This places the most recent reference at the bottom. That’s where you should look for our most recent recommendations.
We’re getting ready for SHARE in Anaheim on February 27-March 4. I’ll be giving my next Hot Flashes presentation that Friday morning. You can still register at www.share.org.
Because this is simply the second of a multi-part series, you won’t find our normal Management Summary on the next page. Tom will return with the next issue.
2. Corrections to Tuning Letter 2011 No. 1
There are two corrections for Tuning Letter 2011 No. 1. On the first page, the heading said 2010 No. 6. That was corrected immediately and resent to all subscribers. Please be sure that you have the correct copy.
On page 21, we included the following paragraph in the z/OS 1.12 IEAOPTxx parameters:
- Added MANAGENONENCLAVEWORK=YES/NO to specify whether non-enclave transaction work of queue servers and enclave servers is to be managed or not. We’ll cover the recommendations for this parameter in our next newslet- ter when we include an article on WLM Updates, which will cover several re- cent updates to WLM. This function is currently exploited by DB2 and you can learn more about it in Horst Sinram’s SHARE in Boston presentation 7777 (z/OS Workload Management Update for z/OS V1.11 and V1.12).
We should have said that the function is currently exploited by WAS instead of DB2.
3. SHARE in Anaheim
We’ll be heading to Anaheim for the SHARE conference that runs from Sunday afternoon to Friday at noon. My Hot Flashes session will be on Friday at 9:30. Hope you stick around. If you’re going to SHARE, you’ll probably have as many difficult choices as I will. For those looking for useful presentations, I’ll give you a list of sessions below especially applicable to z/OS performance and capacity planning. You can probably find me in some of them. Or if you can’t make it to a session, then download the handouts by going to www.share.org, and searching in the Anaheim proceedings. Please note that this is only a small subset of everything available at SHARE. Now you can see why my brain is tired at the end of any SHARE week!
Monday
08:00 8928 Keynote – General Session: Smarter Analytics for Big Data
09:30 8928 Keynote – The Future is Now – Next Generation Mainframe Management
11:00 8678 MVS Core Technologies Project (MVSE) Opening and z/OS Hot Topics
11:00 8857 Workload Management Update for z/OS 1.11 and 1.12
12:15 8249 Lunch Time Brown Bag: 10 Completely Obvious Tuning Tips
13:30 8921 Coupling Facility Nondisruptive Dump
13:30 8798 EWCP Project Opening and ATS z/OS Performance Hot Topics
13:30 8690 z/OS New Facilities and Features
15:00 8940 Getting the Most Out of the z/OS Workload Manager
15:00 8415 Reduce Costs: Getting the Most out of zIIPs and zAAPs with DB2 for z/OS
15:00 8976 What’s New in DFSMS Overview
15:00 8861 What’s New in z/OS Performance
15:00 9019 zNextGen Project Opening and Keynote
16:30 9066 A Mainframe Guy is Still Thinking about Cloud Computing
16:30 8915 RMF: The Latest and Greatest
16:30 8688 z/OS 1.12 JES2 New Functions, Features, and Migration Actions
16:30 9038 z/OS Planned Outage Avoidance Checklist
Tuesday
08:00 9118 Keynote – Eleven in 2011: Top Trends Driving Social Business
09:30 8914 Gathering Workload and Transaction Performance Data using WLM
Service and Reporting Classes Collection
09:30 8920 IBM zEnterprise Technical Introduction
09:30 8206 Network Performance management 101
09:30 8378 Performance Tuning for WAS for z/OS – Practical Advice
09:30 9024 z/OS Basics: The z/OS UNIX Shared File System environment and how
it works
09:30 8695 z/OS Performance Monitoring Tools Shoot-Out: ASG, BMC, CA, Rocket
11:00 8356 DB2 Performance Topics
11:00 8670 IBM zEnterprise 196 (z196) Hardware Overview
11:00 8751 Storage Virtualization with the HDS Virtual Storage Platform: Saving
time, Reducing costs, and Transforming the Data Center to the
Information Center
11:00 8791 Sysplex: Introduction of Coupling Facility Requests and Structures for
Performance
11:00 8696 z/OS Problem Determination Update: z/OSMF Incident Log, Runtime
Diagnostics, PFA, and New Technologies
13:30 8289 CICS Performance Analysis Essentials
13:30 9025 IBM zEnterprise zBX Hardware Overview
13:30 9028 Parallel Sysplex Resiliency
13:30 8792 Sysplex: Key Coupling Facility Measurements – Cache Structures
13:30 8899 z/OS QMgr – Monitoring & Auditing (Including SMF and How to Use It)
15:00 9143 Mainframe Tape Without Tapes – Users Share Their Perspective
15:00 8795 Service Management: The Next Step after the Basics of Performance
and Tuning
15:00 8685 Sysplex Failure Management: The Good-The Bad-The Ugly
15:00 8399 Workload Optimization with IBM Smart Analytics Optimizer
15:00 8922 z/OSMF 1.12 Overview
16:30 8667 CA’s Mainframe 2.0: Simplifying the Installation, Deployment, and
Configuration of CA Products on z/OS
16:30 8297 Changes in Workload License Charges with the zEnterprise
16:30 8604 Latest ISPF Development Enhancements
16:30 8911 Manage and Monitor Your z/OS Unix Services Environment Using WLM
16:30 8698 z/OS 1.12 Implementation and Configuration
Wednesday
08:00 8269 CICS Performance Tutorial–Other Tuning Areas
08:00 8833 COBOL Performance – Myths and Realities
08:00 8764 Getting started in (z/OS) Capacity Planning – Part 1
08:00 9063 How do you do what you do when you’re a z196 CPU
08:00 9035 Smarter SVC Dump Processing for Improved z/OS Resiliency
08:00 8919 z/OS 1.12 SDSF Update
08:00 9022 zNextGen’s A SHARE’d History of the Mainframe: The Chronicles,
Artifacts and Stories!
09:30 9073 Configuring Health Checker for z/OS Hands-on Lab
09:30 8971 DFSMShsm Best Practices
09:30 8923 Everything a z/OS System Programmer Needs to Know to Exploit a
z196 Server
09:30 8674 ISPF Hidden Treasures and New Features – Part 1 of 2
09:30 8916 Optimizing your Batch Window – a multi-discipline approach to
speeding up your Batch Workload
11:00 8668 Detecting Soft Failures Using z/OS PFA: Basic and Advanced Topics
11:00 8763 Getting Started in (z/OS) Capacity Planning – Part 2
11:00 8675 ISPF Hidden Treasures and New Features – Part 2 of 2
11:00 8686 System z Platform Performance Management
13:30 8867 Improve Disaster Recovery and Lower Costs with Virtual Tape
Replication
13:30 8681 Migrating to z/OS 1.12: Part 1 of 2
13:30 9033 PROGxx and LLA enhancements in z/OS 1.12
13:30 8883 zPCR Capacity Sizing Lab – Part 1 Introduction and Overview
15:00 8267 CICS Performance Tutorial – DFH0STAT
15:00 8486 Understanding FICON Performance
15:00 8532 Usage Of zPCR Both In Performance Management And Capacity
Planning Studies – Customer Experience
15:00 9098 zPCR Capacity Sizing Lab – Part 2 Introduction and Overview
16:30 8924 Batch Modernization
16:30 8682 Migrating to z/OS 1.12: Part 2 of 2
16:30 8184 TCP/IP Performance for Dummies
16:30 8864 The All New LSPR and z196
18:00 8697 z/OS Requirements: Influencing IBM Development
18:00 8699 z/OSMF Roundtable Discussion
Thursday
08:00 8863 Coupling Technology Overview and Planning – What’s the Right
Stuff for Me?
08:00 8880 Getting the best out of sub-capacity pricing
09:30 8790 Exploring the SMF 113 Processor Cache Counters and LSPRs
09:30 8790 What’s New with DFSMS ICF Catalog and IDCAMS
09:30 9037 z/OS Basics: Migrating from HFS to zFS and Things to Watch Out For
11:00 9027 A Mainframe Guy Discovers Blades – as in zEnterprise “Blade”
Extension
11:00 8978 Best Practices for ICF Catalog and VSAM
11:00 8627 DB2 Performance Tuning: Where Do We Start
11:00 8757 z/OSMF User Experience
12:15 8298 Lunch Hour Brown Bag: Make Better Technical Decisions by
Understanding LPAR Level Software Charges
12:15 8374 WebSphere Application Server 101
13:30 8514 CICS Explorer Update
13:30 8882 CPU MF – the “Lucky” SMF 113s – z196 Update and WSC
Experiences
13:30 9042 IBM zEnterprise 196 (z196) User Experience
13:30 8385 SMF is Now Mandatory for DB2 V10! SMS Best Practices for DB2
15:00 9097 CA 1 Best Practices
15:00 8361 Query Parallelism in DB2 for z/OS
15:00 8531 Using and Getting Benefit From SMF 113 Records – Customer
Experience
15:00 8512 z/OS 1.12 User Experience
15:00 8694 z/OS Parallel Sysplex: Update
16:30 8786 IBM Smart Analytics Optimizer for DB2 for z/OS: How Does it Work
16:30 8327 Optimized Workload Balancing to z/OS
16:30 8377 Performance Tuning for WAS for z/OS – WAS and WLM Interactions
and Concepts
16:30 8862 To MIPS or Not to MIPS, That is the CP Question!
16:30 8692 z/OS 1.12 Sysprog Goody Bag
16:30 8759 zHPF demystified
18:00 9099 z/OS Ask the Experts Panel & MVS Program Closing
Friday
08:00 8664 A z/OS Systems Programmer Productivity Tool Bag
08:00 8901 Managing CICS with Workload Manager
08:00 8318 z/OS Communications Server Performance Improvements
09:30 8797 Cheryl’s Hot Flashes #25
09:30 8985 Still Spinning: A Case Against Solid State Drives
11:00 8666 Bit Bucket x’28’
11:00 9007 Reclaim Those Empty CAs!
11:00 8851 Understanding The Interaction Of z/OS Workload Manager And DB2
4. SMF Type 113 Records
For two years, we’ve been describing the CPU Measurement Facility (CPU MF) for the z10 and z196 machines, and the Hardware Instrumentation Service (HIS) software component. These two facilities can provide insight to the characteristics of each LPAR that is run on your CEC. Because there is very little overhead, and because the information is so valuable, we recommend that everyone on a z10 or z196 turn on collection of the counters that are written to SMF type 113 records.
You can find our discussions in the following Tuning Letters:
• Tuning Letter 2009 No. 2, pages 8-9 – z10 CPU Measurement Facility
• Tuning Letter 2009 No. 3, page 41 – APAR OA27623
• Tuning Letter 2009 No. 5, page 50-51 – z10 CPU Measurement Facility
• Cheryl’s Hot Flashes #21
• Cheryl’s Hot Flashes #22
• Tuning Letter 2010 No. 4, pages 27-29 – Average RNI MIPS
• Tuning Letter 2010 No. 4, page 48 – APARs OA30486 / OA33052
• Cheryl’s Hot Flashes #24
We also provide several references to presentations by IBM’s Washington Systems Center. We think that it’s critical for every installation to implement this facility for two reasons. The first reason is that IBM uses a new metric called the Relative Nest Intensity (RNI) for doing capacity sizing with their zPCR tool. If you provide the SMF 113 data to zPCR, it can determine the correct RNI, and thus workload MIPS, for each LPAR.
If you’re running MXG, we have provided two report programs to analyze the type 113 records. You can find these freeware programs on our website at www.watsonwalker.com/freetools.html.
The second reason to collect 113 records is to send them to John Burg, at WSC. He and others need as much data from installations as they can get in order to validate their calculations for RNI, and to complete other types of analysis. If you send an email to John at jpburg@us.ibm.com, he can tell you what type of data to send. John has given several presentations on CPU MF, such as the Boston SHARE 7717 session (CPU MF – the “Lucky” SMF 113s – z196 Update and WSC Experiences). John will be giving an updated presentation at SHARE in Anaheim, session 8882. This will probably be available at www.share.org by his presentation on March 3. For additional information on how to collect this new data, please see WSC conference session TC000041.
Stay Tuned!