Oct 12 ’10

This z/OS Is for You! A Look at z/OS V1.12

by Editor in z/Journal

IBM previews a new release of z/OS early every year, announces it later that year, and then makes it available on the last Friday in September. And so it is for z/OS Version 1 Release 12—previewed Feb. 8, announced July 22, and available Sept. 24, 2010. It sounds like business as usual, but is it?

z/OS V1.12 is like no other prior release; it has significant performance, availability, networking, usability, and productivity enhancements that many will find valuable. Together with the new IBM zEnterprise System, it may change the way you do business.


With new VSAM Control Area (CA) reclaim capability, applications that use VSAM Key-Sequenced Data Sets (KSDS) can benefit from improved performance, minimized space utilization, and improved application availability. Such applications include CICS VSAM, VSAM RLS, IMS VSAM, and Catalog processing. In the IBM labs, VSAM workload performance improved by up to 44 percent. Actual results can vary, depending on the degree of data fragmentation. In addition, you benefit from avoiding outages that used to be required to defragment and reorganize this data.  

XML is an increasingly important standard used for data interchange and you can exploit this new standard with z/OS. With z/OS XML System Services, validation parsing performance improvement is anticipated to be between 30 and 50 percent, depending on the amount of data being parsed and schema complexity.

z/OS also introduces a revolutionary new fragment validation parsing capability. By revalidating only the fragments being updated, DB2 10 for z/OS pureXML can avoid the costly revalidation of entire XML documents, which otherwise can take many times longer. As of z/OS V1.10, all validating and non-validating parsing performed by z/OS XML System Services is eligible for the System z Application Assist Processor (zAAP) or System z Integrated Information Processor (zIIP) specialty engines. XML on z/OS is both smart and economical.

With z/OS V1.12, there are significant performance enhancements expected for other types of workloads, too. For example, SVC dump capture time can be reduced by up to 90 percent, though this benefit will depend on the amount of data being captured and how much of that is dumped from auxiliary storage.

Network throughput is improved, too. CPU overhead for applying secure Application Transparent - Transport Layer Security (AT-TLS) will be up to 30 percent less, with the precise benefit dependent on the amount of data transmitted, whether the workload is interactive or streaming, and overall z/OS image utilization. Network throughput for interactive, streaming, and Sysplex Distributor workloads can be improved by using new Open Systems Adapter (OSA) Express 3 Inbound Workload Queueing (IWQ), available on the IBM System z10 and the new IBM zEnterprise 196 server.


z/OS also provides terrific availability enhancements. One valuable function mentioned earlier is VSAM CA reclaim, which gives you the ability to avoid planned outages due to VSAM data set reorganizations. Now you don’t need to take down your applications for VSAM defragmentation, avoiding a task you may currently be doing frequently (e.g., weekly).

You can also avoid additional potential outages with the z/OS Predictive Failure Analysis (PFA) capability. With PFA, z/OS learns from its own environment and tracks system trends, predicting system abnormalities and alerting you to potential problems before they occur. You can use PFA on live systems or even earlier, during application development and testing, to identify previously unknown potential problem areas and respond before they become serious issues. If you can address sick systems faster and more accurately, you can preclude a small incident from cascading into a larger one.

If a system problem occurs, a new base component, z/OS Run Time Diagnostics (RTD), is designed to quickly analyze key indicators on a running system and help identify the root causes of system problems and degradations. RTD does what an experienced operator might do, but in a fraction of the time. It will run in as little as one minute to return results fast enough to help you choose between alternative corrective actions quickly and effectively.

Parallel Sysplex is many clustering solutions in one; it’s used for scalability, availability, software migrations, and disaster recovery. It, too, will benefit greatly from enhancements in z/OS V1.12. The Sysplex Distributor (which manages Sysplex TCP/IP traffic) now has a hot standby capability. You can configure a preferred server and one or more hot standby servers with automation that can switch between the two. In the event of a Coupling Facility (CF) incident, z/OS can now capture CF data non-disruptively, allowing the CF to continue operating and helping to improve Parallel Sysplex availability. This benefit is available with z196 servers and CF control code Level 17. You can also secure your TCP connections easily with less CPU consumption using the new trusted connections capability.


This release of z/OS boasts impressive productivity improvements. The new z/OS FICON Discovery and Auto Configuration (zDAC) can simplify configuration of FICON-connected disk and tape. zDAC can automatically discover new or updated devices. It can then propose channel configurations based on high-availability best practices, existing configurations, and even taking into account Parallel Sysplex symmetric configurations. This function can save you hours of configuration time and is available only with the new IBM zEnterprise 196 server. zDAC works with any FICON-attachable, fabric-attached storage (disk or tape) that registers with the FICON name server.

The z/OS Management Facility, the new face of z/OS, can shave hours off system management tasks. With its embedded guidance and wizards, reflecting built-in best practices, new systems programmers can be productive in weeks, not years! It has never been easier to:

Taking Advantage of a New Dimension of Computing

z/OS exploits the new IBM zEnterprise System, a revolutionary direction in multi-tier, multi-system, multi-workload management and coordination. zEnterprise is beneficial for all System z workloads, but it’s especially ideal for z/OS workloads. zEnterprise provides terrific scalability, availability, and performance for z/OS workloads while improving integration with and management of distributed environments—bringing them close to your core z/OS data.

z/OS is designed to integrate seamlessly with the new zEnterprise System. At a high level, the zEnterprise System is comprised of the IBM zEnterprise 196 server (z196), the IBM zEnterprise BladeCenter Extension (zBX), and the IBM zEnterprise Unified Resource Manager (zManager). Most z196 function is available on z/OS V1.10 (with support as far back as z/OS V1.7). Some z196 functions, such as zDAC and the OSA Express-3 IWQ, are exclusive to z/OS V1.12. Only a handful of new networking configuration settings are needed to connect z/OS to the zManager. A new agent in z/OS V1.12  enables you to send basic z/OS WLM information to the zManager. Also, DB2 9 for z/OS (part of the IBM Smart Analytics Optimizer for DB2) is the first resource that exploits the new zBX class of optimizers. The IBM Smart Analytics Optimizer can send CPU-intensive queries to execute on the zBX, returning the data warehousing results in a fraction of the time.

Customers have found in early tests that z/OS, combined with zEnterprise System, can provide:

z/OS V1.12 strengthens its core value by delivering advanced real-time capabilities designed to provide higher performance, improved productivity, new levels of data availability, and fewer disruptions to z/OS and the business applications that rely on it. The zEnterprise delivers new management and control capabilities to the servers and technologies that rely on z/OS. Together, they represent a new era of management, monitoring, and resiliency for your critical multi-tier workloads.