Wednesday, June 30, 2010

Looking at Autorefresh

We've discussed at length DASD monitoring considerations, and the cost of monitoring when it comes to DASD performance metric collection. What I would like to do next is take a look at other aspects of the cost of monitoring, and take into account such things as optimization of CPU required for monitoring.

One of the first things to look at when it comes to looking at the cost of monitoring is if you and your users are employing Autorefresh. Autorefresh implies that OMEGAMON will be regenerating a given monitoring display (screenspace, CUA screen, or TEP workspace) on a timed interval. In Classic and CUA interface, Autorefresh is set in the session options, and if used extensively, Autorefresh can measurably drive up the the CPU usage of the Classic and CUA tasks. For example, if you have multiple users, each running a display on a relatively tight (10 seconds or less) interval, OMEGAMON is doing a lot of work just painting and re-painting screens on a continuous basis.

The recommendations are as follows:
Limit the use of Autorefresh
If you must use Autorefresh, set it on a higher interval (60 seconds or higher)
Better yet, if you must use Autorefresh, us the Tivoli Portal to drive the display. The TEP has more options to control Autorefresh, and you will be moving some of the cost of screen rendering from the collection tasks to the TEP infrastructure.

Thursday, June 24, 2010

Webcast on System Automation V3.3

Sorry for the short notice, but there will be an interesting webcast on June 24th on "Achieving high availability and resiliency for your business with the new capabilities of Tivoli System Automation Version 3.3". If you can't make the event, you can always download and listen to the material later.

Here's the URL to register and attend:

Broadcast date/time: June 24, 2010, 11 a.m., EDT

The new Version 3.3 of Tivoli System Automation extends it's industry-leading capabilities through increased integration with Tivoli business management and monitoring solutions, additional automation controls, and simplification of automation activities.

In this session, the discussion will include how IBM Tivoli System Automation solutions can:
Provide remote management
Alerting and escalation as an adjunct to automation and high-availability solutions
Leverage automation technology for service management and business continuity

Speaker: Allison Ferguson, Automation Solutions Product Manager

Many thanks and keep stopping by....

Since January we've had over 1000 first time visitors to this blog. The traffic has been very good, and I appreciate all the interest.

I will continue to post technical content, and upcoming events. I have quite a bit more to cover just in the area of OMEGAMON optimization.

In the future I will be doing more to make this blog more interactive. Feel free to post comments. Also, if you have a topic of interest, please let me know. You can put it in a comment on this blog, or you can email me at

thanks again

Wednesday, June 23, 2010

Upcoming IBM webcast and virtual event

On July 22nd from 12 to 2 PM ET, IBM will be holding a virtual event on powerful new technology. Today's IT environments are built using diverse computing technologies x86, Unix, Linux, and z/OS. On July 22nd IBM will be unveiling a solution designed to integrate enterprise systems, people and processes - so they work together as one.

In the virtual event you will have the opportunity to:
- Be among the first to receive the latest technology breakthrough updates
- Network with your peers and IBM subject matter experts
- Participate in online discussions
- Download the latest whitepapers and brochures
- Meet the experts
- Live discussion with Q&A

To attend the virtual event go to the following URL:

Friday, June 18, 2010

OMEGAMON Storage ICAT configuration considerations

When you are installing and configuring OMEGAMON XE For Storage there are quite a few options in the ICAT installation panels. Most of these options are fairly straight forward (such as sampling intervals for tape, cache control units, etc.). But, as I've mentioned in earlier posts, the most critical options revolve around the specification of the DASD collection options.

There are several options that control DASD collection, and there are some subtleties to the optimal way to specify collection. In the example shown here, Option 2 is where you can specify DASD collection frequency, and Option 3 is where you can specify devices to exclude from monitoring. If you can exclude some devices from ongoing monitoring, it may be a good idea to do so, and save the cost of monitoring.

Option 2 (DASD Device Monitoring) is the most critical panel to review. When you set up collection options here, you can either collect information every nth I/O (sample count) or collect when the device exceeds a specified MSR (millisecond response time) threshold. The example screen shot I show, has the default configuration panel. What you get by default is every I/O will be monitored, so my first recommendation is to not just automatically take the default. Take the time to configure the options on this panel.

First you can create an entry that will specify a VOLSER or device range. Once you have specified your devices, you then have a choice of monitoring I/O based on sampling (every n I/Os) or monitoring the device if a specified MSR threshold is exceeded. In the Monitor Status field, if you specify ON, then you will monitior based on the number specified in the Sample cnt/MSR field. But, if you enter MSR in the Monitoring Status option, I/O monitoring will happen if the device exceeds the MSR threshold. The Sample cnt/MSR field is treated as an MSR level setting, and the other part of the magic is that the MSR/exception trip count specified in the top portion of the panel comes into play.

Perhaps a little confusing, but obviously these options make a big difference in terms of how much data is gathered on an ongoing basis. If you want to keep the cost of collection down, then consider using the MSR option to collect when exceptions occur. If you want more data on an ongoing basis, then consider the sample approach, but set the sample frequency higher than the defualt of 1 (which means collect everything).

Wednesday, June 16, 2010

Upcoming Tivoli Enterprise Portal webcast

I will be doing a webcast next month on "Top 10 Problem Solving Scenarios using IBM OMEGAMON and the Tivoli Enterprise Portal". The presentation will look at what I consider to be one of the strengths of the Tivoli Portal, its power as a problem isolation and analysis tool. We will look at examples from all the core OMEGAMONs.

The webcast is July 22nd at 10 AM Eastern time. It's a free webcast, so the price is right. Here's a link to register and attend:

Monday, June 14, 2010

New OMEGAMON configuration methods

For years OMEGAMON users have installed and configured the product using an ISPF based dialog called ICAT. ICAT has been the primary mechanism to configure and manage all the core OMEGAMON tools (OMEGAMON for z/OS, CICS, DB2, IMS, Storage, Networks).

Over the past few months a new configuration method, called Parmlib, has been rolling out. Phase 2 of Parmlib support became available in May. Parmlib is intended to be an easier to use alternative to the ICAT process. In the June issue of IBM System z Advisor, Jeff Lowe has written an article that talks about Parmlib. If you are responsible for installation and configuration of OMEGAMON, you will find this article of interest.

Here is a link to the article:

Friday, June 11, 2010

OMEGAMON Storage - Application level monitoring

OMEGAMON XE for Storage offers the flexibility to monitor DASD and I/O at the level of the application. What this means is that you can use the tool to focus your I/O monitoring strategy on the most critical applications, without having the need to monitor the entire DASD complex.

The example shown here shows how you can enable Application level monitoring. To enable application monitoring you need to create a situation to invoke collection. The primary parts of the situation are the application name (which can be wildcarded to include multiple tasks), the monitor switch (Y/N), and monitor I/O option (which controls if you want data set level monitoring). Given these options you have the flexibility to control what tasks you want to gather I/O information about, and whether you need data set level I/O detail.

The picture here shows an example of how to specify the situation to control application level monitoring.

Thursday, June 10, 2010

OMEGAMON Storage - User DASD Groups

OMEGAMON XE For Storage has many collection options for gathering DASD performance and space management information. User DASD groups are one option I find most interesting in that it gives the user considerable flexibility in terms of how DASD data is gathered and displayed. This flexibility is useful in terms of being able to control the cost of DASD monitoring.

A user DASD group is a user-defined logical grouping of DASD volumes related according to specific criteria. You can define a user DASD group based on volume name, device address, SMS storage group, or one or more aspects of the DASD device (examples include fragmentation index and MSR time). Once the user DASD group is defined, you can use the User DASD Group Performance and User DASD Group Space workspaces to view aggregated performance and space metrics for the group.

User DASD groups can be defined in a couple ways. One is via member hilev.RKANPARU(KDFDUDGI). The KDFDUDGI member allows for the definition of a group based on volume, a range of devices, or SMS storage group name.

Another way to create a user DASD group is via the Tivoli Portal (assuming V4.20 Interim Feature 2 is installed). In the TEP select the User DASD Groups Performance node on the navigation tree, right-click any row on the table view, and select Add Group from the menu. As before, the user may specify constraints by volume, device address, device range, SMS storage group, or volume attribute. There is also an attributes tab to specify the attribute constraints that are used in conjunction with the DASD device constraints specified. There is a drop-down list in the attribute column to specify the attribute information to be used.

I like user DASD groups because it provides a way to control monitoring options for various types of devices. Device architecture and usage varies in many enterprises. Being able to group devices makes the analysis more meaningful for a given environment.

Friday, June 4, 2010

Access System z from a Smart Phone

If your business is looking at exploiting technologies like the iPhone or the Droid, fear not, z/OS can work in this environment. Here is a link to an interesting IBM Red Book that goes into how it's possible to combine the traditional strengths of the mainframe to manage large volumes of data and run business transactions with the Web 2.0 paradigm. And to take advantage of simpler interfaces, better integration among different services, lightweight protocols for communication, and much more, together with the availability, security, and reliability of mainframe data.

Here is the link to the Red Book:

Thursday, June 3, 2010

More on DASD monitoring - OMEGAMON XE For Storage

We've discussed DASD monitoring and the cost of monitoring for several of the OMEGAMON solutions. Now lets consider OMEGAMON XE For Storage, and the cost of monitoring considerations for this solution. As the name implies, OMEGAMON XE For Storage is an OMEGAMON solution focused on the monitoring of DASD on the z/OS platform. But, OMEGAMON Storage provides more than just DASD monitoring. OMEGAMON Storage also provides information on virtual tape, channels, control units, cache controllers, HSM activity, shared DASD, and even more. Another thing to be aware of is that OMEGAMON Storage has been evolving over the years, and has morphed into a much broader storage management, as well as a monitoring tool. OMEGAMON Storage now includes integration with a array of Tivoli Storage solutions, including HSM and backup tools, via the Tivoli Enterprise Portal.

Referring back to a primary consideration I mentioned in earlier posts, the more data you ask for, the more potential cost from a cost of monitoring perspective. OMEGAMON Storage certainly falls into this area. OMEGAMON Storage provides the ability to monitor storage at many different levels: shared DASD, the individual LPAR, the channel level, the controller level, the UCB level, even down to the individual dataset level. The user also has the ability to monitor DASD, I/O, and space utilization from the perspective of the workload applications on z/OS, and also to make user defined groups for monitoring and space analysis.

Clearly, OMEGAMON Storage provides useful and detailed information. It is important that the user have a well conceived plan when deploying OMEGAMON Storage to avoid redundant monitoring cost. When installing OMEGAMON Storage, the tool is usually installed on each LPAR in the z/OS environment. In a shared DASD environment, probably the first recommendation is to have the OMEGAMON Storage instance on a specified LPAR in the shared environment be the primary collection point for DASD information, thus avoiding redundant collection of DASD information on multiple LPARs.

There are quite a few more considerations for the set up and optimization of OMEGAMON Storage, and I will get into these in more detail in later posts.