SYSVIEW Performance Intelligence Reviews

For UPS, we are trying to match up end-to-end between distributive and the mainframe. So far, the AutoSys view, it looks good on a mainframe. We are just having trouble matching it up on the distributive side, but the data collection and everything looks fine to us mainframe folks.

The benefits are kind of what we are looking for. We are not there yet, and end-to-end, we are trying to tie it up with our distributive apps to match them up with the mainframe apps. Therefore, everything is tied through one single pane of glass.

Like a Megamind does, it provides CPU contention batch job information. We are looking more for our kicks regions and stuff, which it self discovers, then they have some work to do. However, it has been very valuable so far.

The product itself seems very good. The documentation, like installation guides and stuff, put it all in one place.

We are still in early stages. Right now, it looks fine from what I see of it. We have only really put the CISS regions on it. We have not pushed out anymore online or anything yet.

Yet to be determined.

Between the tech support and the person I was working with, they have been extraordinary.

I was surprised at how easy it was to work with them. I have worked with them in the past and I did not always find it that way. In the past, I am going back about 10 years. Now, they seem to have come a long way.

We were and still are using Megamind. We still are because it has been customized for our online regions, but it does not give us the end-to-end. So, that is the reason we are going for the SYSVIEW and the APM, as well.

It was dictated to me from upper management. Apparently someone from CA spoke to one of our upper management. It was the end-to-end solution to what we have been looking for at UPS.

It was a little tedious, but it was brand new to us. The documentation kind of jumps around because there are all different parts. However, the support has been outstanding.

There was a lot to do and different multiple pieces to it. It is not just one piece because of the way a SYSVIEW operates, touching almost every part of the operating system.

We knew it was coming like that. Just being the only person, it takes a lot of time. We do not really have a direction set for what we are trying to accomplish, so that makes it a little tougher.

So far, I am the in-house team.

Compuware and another vendor, but CA came in with the best deal.

You need to be partnered with CA. Work alongside of them. It makes it a lot easier because of the amount that goes into installing this product. You will need a little help.

Most important criteria when selecting a vendor: usability and scalability.

Actually, all the features are valuable. The product gives deep insight into a z/OS operating system by means of many powerful commands, each of which provides information within seconds. By contrast, using normal system commands would take minutes, or sometimes even hours, to get similar information, if at all. Similarly, there are many powerful commands for modifying a z/OS operating system.

I can emphasize this with the following examples:

1. Instead of using z/OS system commands on a system console in order to get some information from the system or to modify the system, you can use CA SYSVIEW commands to do exactly the same. Compared to a system console these commands provide a tremendously more user-friendly interface to the system and the output is being displayed much more clearly arranged and accessible than console output. This means much more efficiency, e.g. for system operators.

A specific example of this is all the Dynamic Linklist commands you can use to display and modify the MVS linklist concatenation.

2. There are a lot of commands providing system information as well as system modification for which there is no similar native support from z/OS, especially in a case of emergency. This means that, without a product like CA SYSVIEW, you either had to write your own programs providing such information/modification or you even had no chance at all to deal with such a situation. This means high productivity, e.g. for system programmers.

A specific sample for this is the ACTIVITY command with all its line commands you can use to display and modify all running address spaces in various ways, e.g. displaying allocated data sets, active enqueues, task structure, canceling address spaces of even individual tasks.

Several times it saved us tens of thousands of Euros by tremendously reducing the time necessary to solve a severe operating system problem.

As with every piece of software, there’s always room for improvement. But for our shop, the most important features of the product have been working perfectly for a very long time.

As for the JES2 component of the product, CA might improve the overall performance when accessing the JES2 spool. I do know that CA is using system interfaces from IBM to get the JES2 data, and I suspect that the actual room for improvement is in that interface. Maybe CA cannot do very much to improve performance, but it would be great to get better performance in this area.

We have been using this solution since 1993.

There are rarely stability issues, just minor issues when they do occur.

There have been no scalability issues.

In the 90s, the European first-level support was sort of lukewarm. This changed positively with the beginning of this century, and it still persists.

We had a previous solution, and we switched because CA SYSVIEW was clearly superior.

The installation was, and still is, straightforward, provided that you are a moderately experienced z/OS systems programmer.

The only additional candidate then was OMEGAMON from IBM.

A great deal of the product is definitely suitable for z/OS system specialists with some years of job experience. In other words, z/OS beginners will probably not be able to appreciate all of the product’s features. Thus, it should be evaluated by an experienced systems programmer.

The most valuable thing to us about SYSVIEW is the overall scale of it. They have a lot of jobs with a number of conditions that go into the overall job management. This gives it the ability to scale and deal with various job dependencies.

It's intuitive to use and you don't need a lot of expertise, but you need to know how to manage it.

I think some of the messaging from SYSVIEW could be improved. For example, how do you know if there’s a real issue that needs to be addressed? Is there something that identifies a problem in your configuration? Is there something for which you may need pro services?

While we do have consoles, what we attempt to do is to get those alerts integrated or reported into our monitors, and it wasn’t a very clean integration. We tried things like syslogging, but this required a lot of massaging and it's still not perfect.

After the complexity of the initial implementation, we've had no issues with deployment.

The system is extremely stable.

When making that call to technical support, the triage is sometimes a little rough to get you to someone who will effectively help you. When trying to solve some performance issues, it was better for us to engage and get them side-by-side. There were times when we had to go outside of CA with known experts.

We migrated from our older solution to this one. This is much easier to manage. Our previous solution was heavy with a lot of baggage which required a lot of resource commitment and getting experts in.

The initial setup was complex and accompanied with the fact that we didn’t have a lot of latitude. Also, our workload manager gave us added complexities, so we didn’t get to do a lot of greenfield stuff, requiring a bit of massaging.

I would say first and foremost, understand your business rules and understand what you are trying to accomplish and your requirements to the fullest. Then make sure they are articulated so you can make a good decision. A lot of time, folks look at a product pipeline and not looking at their requirements.

• IOF/SDSF compatible commands and displays
• MVS displays and commands
• DB2
• USS CICS displays

Thresholds with alerts are valuable, enabling us to fix a problem before it has an impact.

The MVS commands, because we have a lot of new, inexperienced users on the mainframe who don’t know the commands.

The documentation. I noticed with the last release that all of the manuals were bundled into one. It used to have a separate installation guide, user guide, etc.

No stability issues, it has always been reliable.

No issues with scalability.

Excellent.

SDSF and IOF. We didn’t switch, we have both products.

Very straightforward, with excellent installation documentation.

I am not involved with pricing or licensing.

SYSVIEW has been here since I started working here.

Follow the manual and you won’t have any problems. If you do come across any installation issues, instead of spending lots of time trying to resolve it, open up a question with support and they will resolve it very quickly.

We're using CA SYSVIEW to monitor performance and availability on the mainframe. We replaced MainView by BMC with it. It was a cost savings for us.

It integrates very nicely with other products from CA, like OPS/MVS, and it's easy to learn.

There are some parity features, things you can do in a JES2 or JES3 shop, which you can't with CA SYSVIEW. For example, purge output from the spool. So better parity between JES would be useful.

It has been very stable. We've only been using it for a couple months, but we haven't had any problems with it.

We're a pretty small shop, so I'm sure we can scale it up to what we need it to do.

Technical support is very good. They get back to us very quickly and usually have a very good answer. Even if it's one I don't want to hear, but they're good about that.

We previously used MainView from BMC. Our contract was up for renewal, and the sales lady offered CA SYSVIEW as part of the contract renewal. The pitch was cost savings, and it was priced better, so we went with that.

For us, vendor is general. It's mainly the product that we care about, but cost is definitely in there. This does pretty much the same thing that MainView does at a cheaper cost. That's why it won.

I installed CA SYSVIEW. The actual product install is very simple. The web interface install is far from easy. The manual isn't very straightforward on that at all because it's pieces of a different product. Installing it on web interfaces was a challenge.

We didn’t evaluate any alternatives for this product.

Just take your time and set thresholds and alerts properly.

The CICS option is used very frequently. For myself, I mostly use the z/OS part. I also really like the XSYSTEM functions in case I'm having trouble logging on to an LPAR.

We have automated the monitoring of our CICS regions and CICS transactions. This leaves much less manual monitoring for the CICS people than before.

The creation of dashboards could be made much easier, such as the way it is done in BMCs MainView for example. Currently, you need to do some programming to create dashboards, which is time consuming and error prone. A dashboard you can create in seconds in MainView takes much, much, more time to create in SYSVIEW.

We have never had any issues with stability. The product runs without any problems. On the rare occasion that there is a problem, the recovery is really good, so there was never an outage of SYSVIEW at all.

We have had no issues with scalability. Additionally the product needs fewer CPU resources than other products.

The technical support is very good. You always get an immediate response as to who is taking care of your issue, and you get regular feedback of what’s going on. Normally the cases are solved in a much shorter time frame compared to competitors.

We previously used Omegamon. I think we switched because of financial reasons. Probably CA made a quite good offer, but this was before I started here, so I don’t know about the details.

The setup is not really complicated, but there are many configurations options, where you can easily get lost.

I can’t say anything about pricing and licensing as I’m just the techie and have nothing to do with contracts.

The decision for switching to SYSVIEW was made long before I started here.

Start with a small and easy configuration and get it running. Make yourself familiar with the product, then switch on additional features one at a time and get familiar with it. Otherwise, you may get lost in options, features, and parameters. The product is quite comprehensive.

Almost all of the features are valuable but specifically z/OS (base), CICS, IMS and MQ.

The TCP/IP feature is used a little, and the CAPTURE feature occasionally, helping a lot when dealing with some issues.

One of the most important usage for us is implementation of alarms based on product-provided metrics (especially on loops). In CICS monitoring area, we even use autocancel functions based on threshold settings.

Perhaps the IMS monitoring part could be improved (at least according IMS specialists).

I have been using it for over three years.

I have not encountered any stability issues. Most issues have been minor errors or performance issues (most of those were handled by appropriate parameter settings).

I have not encountered any scalability issues; setting is quite granular.

Technical support is very skilled and helpful.

I previously used the BMC MainView family of products. We switched to CA mostly for cost reasons.

Initial installation and setup is quite straightforward but after that, some customization and planning is necessary if you want to get the best out of the product. But the customization part (or parameters for customization) is quite well documented.

Before choosing this product, the IBM solution was another option under consideration.

In previous releases, the default setting was consuming too much CPU perhaps, but since 14.1, this was tuned quite well. Still, I would advise checking the interval setting in SCHD* parmlib members in case you want to keep CPU usage by SYSVIEW as low as possible. Using ZIIP is also good thing, where applicable.

System monitoring and issue alerting. Problem investigation and resolution. System status and configuration details.

It pulls together a lot of utilities provided by other tools/products into one consistent umbrella.

CICS, JES, and MVS aren't necessarily valuable, just very useful. SYSVIEW provides information that supports our duties in problem alerts, problem investigation, and performance monitoring and management.

For one particular environment, the capturing, recording and presentation of IBM CPUMF data would be useful.

IBM Mainframes now produce data on processor cache performance. There is a mechanism for capturing that data within z/OS. It is useful for comparing machines and the impact of the number of partitions and logical to physical processors ratios.

We've had no issues with the stability.

It's been able to scale for our needs.

They're very good.

We switched due to a corporate directive.

It wasn't very difficult to implement.

I am not involved with licensing, as this is done at a corporate level.

I was not involved in the decision to purchase this product.

It is a very powerful and reliable tool with a relatively easy to follow interface.