Latest Entries

Understanding FICON Channel Path Metrics

While FICON channels offer significant advantages in terms of channel bandwidth and performance, many users long for the old steam gauge utilization metrics provided for ESCON channels. While there are no simple rules-of-thumb for FICON channel management, this article will explore the available path utilization metrics and suggest new metrics for FICON channels.

Read Full Article →

Picking the Best SQL Statement to Tune

You may have read the title and are now thinking to yourself that the best SQL statement to tune is the one the users are complaining about today. I would not argue with that perspective. However, I would like to present a different approach to SQL tuning. This discussion will not take the place of firefighting problem transactions, but it just might remove those situations from your routine. Proactive performance tuning has been discussed for years. This article examines a methodology that can be applied to any application for purposes of tuning the portions of the application that would provide the greatest return on that tuning investment. It is based on the Pareto principle, or the “80/20 Rule.” When applied to a DB2 application, this rule presumes that 80 percent of the resource usage (CPU, memory, and I/O) is being consumed by 20 percent of the SQL in that application. Therefore, tuning that 20 percent of the SQL statements would provide the greatest ROI…

Read Full Article →

Most performance tuning activities involve systems tuning such as managing I/O, adjusting parameters, moving workloads, etc. Only rarely do you work directly with application developers to improve code efficiency. But there’s a third tuning method—one that’s only rarely employed, but often can be highly effective. This is particularly true when you can’t touch the application…

Read Full Article →

Reading in the last issue of z/Journal about plans to set up a Mainframe Executive Forum, I’m reminded yet again how essential it is for mainframe operations and technical managers to meet on a regular basis, not just to exchange news and views, but to help retain a voice within the broader IT community. Internet discussions are great for exchanging technical tips, but there’s really no substitute for the physical shaking of hands and meeting of the minds!…

Read Full Article →

Storage networking devices include processors (host servers), channel and host bus adapters (HBAs), switches (including directors), and storage devices (disk and tape, including virtual tape). Some additional storage networking devices include bridges, gateways, and routers, which were mentioned in the October/November 2003 z/Journal article, “Has Open Storage for zSeries Finally Arrived?”. Figure 1 shows a storage network with two separate fabrics (I/O paths) for redundancy (Path A and Path B) with servers attached to both. Figure 1 also shows servers that are single-attached only to Path B. Servers are attached to switching devices (director or switch) that have storage attached to them. Figure 1 also shows a switching device connected to another switch on the right using interswitch links (ISLs) in what is known as a cascade topology. This is an example of a metropolitan area network (MAN) or wide area network (WAN) for remote storage access and mirroring for data protection and access…

Read Full Article →

Exploiting CMS & CP in z/Linux

As we’ve seen in previous articles, Linux on the mainframe is largely similar to Linux on a PC. However, the main thrust of this ongoing series has been to show how Linux guests in a z/VM environment can leverage VM’s capabilities to do things otherwise impossible on a stand-alone Intel server. In this article, we’ll examine the use of CMS as an extremely smart BIOS for the z/Linux guest and how to maximize CP services and the CMS file system from within Linux itself.

Read Full Article →

LINUX For zSeries: Expectations vs. Reality

It has been four years since IBM Distinguished Engineer Dr. Karl-Heinz Strassemeyer talked at length about running Linux on the mainframe at the World Alliance of VM and VSE (WAVV) Conference in Fort Mitchell, KY. Linux on the mainframe has come a long way since then. There were—and still are—many high expectations of the platform, which have been tempered somewhat by a big dose of reality in the form of deployment successes and failures. It is easy to talk about the successes, which range from the miniscule to the downright glorious. However, it isn’t as easy to talk about the failures, which at times threaten to cast a pall on the platform and spawn fear and doubt for the new Linux on zSeries technology…

Read Full Article →

zHoliday Storage Spirit

From FICON to fabric and “virtualization” to the many “natural” and manmade events that show disaster recovery in action, what a year it has been! Now that the holidays are upon us, I thought it would be a good time to reflect on the year gone by and present my wish list and predictions for the year ahead. So, what’s on your wish list this year? More storage, disk, or tape? There’s connectivity and interfaces for FICON and more switch and director ports. How about we dust off last year’s plan and finally put in that SAN? What about LANs, MANs, WANs, SANs, POTs (pretty old technology) and PANs (pretty amazing new stuff)? Is this real or part of the “virtual” storage land?…

Read Full Article →