(Road) Rage against the machine: Understanding - and dealing with - z14 traffic patterns
The z14 has vast throughput potential. With almost 200 Processing Units (PU's) and 32 TB of cache memory, this is a box designed for massive, parallel processing. So why do the same jobs sometimes take longer? Why do I need to turn on additional engines when transaction rates seems similar? Just as higher speed limits or more lanes don't always solve traffic problems, more MIPS or more memory is not always the answer.
In this session, John will share his insights from over 25 years of performance analysis. we will explore common sources of application delays and discuss practical solutions to reduce these delays. Stop the rage. Don't get mad; get moving.
Time: 09:00 - 10:00
John Baker, Senior Consultant at IntelliMagic, has been working with mainframes for over 25 years, both as a customer and consultant. His primary focus has been on system and application performance. As a customer, John designed, implemented and maintained many critical projects such as WLM Goal Mode, GDPS/Data Mirroring, and merging datacenters. As a consultant, John has assisted many of the world's largest datacenters with their z/OS performance challenges. John is a frequent and much appreciated speaker at industry events like SHARE, CMG, GSE and IBM conferences. He has served on the Board of Directors of the Computer Measurement Group (CMG) and currently volunteers with the MVS Performance Project (MVSP) at SHARE. In his spare time, John enjoys pool, motorcycles, his dog, old port and a spirited political debate. He lives with his family near Toronto, Canada.