CA NetQoS Performance Center [EOL] Reviews

CA Application Delivery Analysis (ADA)
When it comes to application and client slowness, this module is able to monitor and analyze TCP packets between clients and servers, providing network metrics, latency, TTL and more. It comes with a friendly GUI, filter, and report function, is easy for management level to understand, and helps network engineers narrow down the problem area.

CA Network Flow Analysis (NFA)
This module gives you both live and historical network traffic data. Shows which application IP, protocol, or client IP is using the bandwidth, or whether it is an ISP issue.

CA NetVoyant
Some applications are very sensitive to latency, jitter and bandwidth (they cannot mix with other traffic). This module helps by providing device status, device metrics, and device utilization.

CA Event Manager
For critical processes that we may need to monitor 24/7, this module helps collect alarms, traps and log streams. This enables the operations team take action quickly.

With this product's help, it can greatly ease:

1. Troubleshooting slow issues based on flow data.
2. Troubleshooting high bandwidth utilization and tracking down the root cause.
3. Bandwidth usage reporting for bandwidth capacity planning.

Reporting. Reports can take a long time to generate. The report format types also need to be increased.

I used this product for four years.

Nothing is perfect. You can refer to the CA knowledge base and their forums.

Yes. This product needs a super good server in hardware and bandwidth, when there is growth in usage.

Eight out of 10. Overall, the technical support provides good service.

Yes. We switched because the product's overall price and technical support level meet the company's requirements.

The setup is very straightforward.

Price and licensing are reasonable and acceptable.

Yes. SolarWinds and ManageEngine.

If you need solid data and reporting for management, planning, and troubleshooting, this could be the right product.

I think it's a good product to pull data together in an easy-to-search-through fashion, instead of having a lot of different individual solutions. It provides SNMP traffic statistics instead of alarms, which Spectrum provides instead.d

It gives us an easier way to start troubleshooting problems. It lets us dig more granularly, so if we saw something come in a couple hours ago the old way of looking at the stats was very, "This is what you're going to get," but now you can dig down and say, "I want to see the specific time frame," or even "I want to dig back days ago and see a two-hour block of what was going on and try to isolate what caused it."

We've started getting into moving modules around and kind of designing what we're seeing at a given instance. So, if we know that when we troubleshoot a certain type of outage, we like to see this data, we can present just that data on a dashboard, then that's powerful. That would save a lot of time instead of having to dig through all the data. So, I guess a more organized view of all the data that it provides would be an improvement we'd want.

We've had no issues with deployment.

We've had no stability issues with the product itself. We may have put under-powered servers running the application or things like that, and those are some issues that we're working on internally to actually make it supported by CA. A lot of it really has to do with how much time you invest in setting it up and having the right hardware.

We've got more devices than what should have been on certain data pullers. Once we got that upgraded, and we did upgrade a couple months ago to the new versions, everything just kind of came together where we could see that performance all of a sudden was like, "Okay, this is exactly what we were hoping it would have done."

We haven't needed to really make a lot of calls to technical support on this side.

Other SNMP pullers that just give you very static info, you might see a whole day's worth of traffic in one graph. You can't dig down further into that. You might have a harder time seeing statistics for memory uses or anything because you've got static, defined time frames. You can see it, but you can't see it, if you will.

It's probably fair to say that I think we probably don't use all of the features that are there already, so as we get more and more into what it can do, it's met everything that we've wanted at this point.

To be honest, we have not really had a lot of experience with this tool. We sort of installed it to meet a customer request requirement. It was an early version of the product about almost four years ago now. It didn't get much focus beyond that initial deployment. Now there is a renewed high interest in the product, so we have upgraded to the latest release.

We understand that the product still has some growing pains, but we are impressed by the ability to integrate that product with Spectrum and with other tools. One of the things we want to focus on here is understanding the APIs for both Spectrum and for Performance Center so that we can more tightly integrate those products with other tools that we use.

One of the key benefits is doing the performance management piece where you can crack the utilization of network connections and see where your block points are, and if you have errors happening that effect performance that you don't necessarily see as alarms. You can set thresholds inside the Performance Center for those alarms.

I'd like more ease-of-use. Although that is improving, and it is with each successive release, that aspect does need improvement. We like to turn over, as much as we can, the administrative tasks to end users so that they don't need to call us to make a change to their layout, or their polling intervals, or whatever. It's great having the ability for someone with minimal training to be able to do that. Those kinds of things on their own are really important.

I don't have enough experience even though we've had it up in the environment for a while. Although we've not really had any issues with stability in terms of keeping things up and running. We haven't really been beating on it very hard either.

You simply add collectors to do the same footprint. The collectors do the polling and the gathering of data for performance metrics. It has the same capability of scaling in that regard.

I haven't really had as much experience with that.

I didn't have a direct role in that, but no real issues. It's a more complex solution in terms of a separate database server, an aggregator that talks to all the collectors and the GUI, and the application service.

The most valuable feature I think is the ability to go in and perform updates yourself. You can change around any of the parameters and different types of dashboards you want. If you want bar charts or heat maps or anything like that, you can go in change the different graph types.

I also like some of the canned dashboards that come out-of-the-box -- the infrastructure, server, and services overview.

We use it for monitoring our network, MPOS, and top-trending interfaces that are either over-utilized or under-utilized. We're then able to change the bandwidth of those accordingly and save money.

It also help us with the back-end stuff, such as authentication as we have people all over the world trying to log on.

It would be really nice if you could do multiple domain authentication as opposed to just a single domain.

CA Performance Center is relatively new for us.

We have 40-50 users right now, though we'd like to get it out to all our end users and to all of IT so that they, too, can monitor their stuff. But until we can have some type of standardized authentication or multiple-domain authentication, then it's not likely that we'll scale more than that.

We worked with John Wittie until he retired from CA. He was absolutely essential in getting it installed cleanly and up and running.

We never really evaluated other options, but we did look at EMC Smarts and IBM Tivoli.

You should start slow and have a definite scope of work that you need to get accomplished. Pick one thing that you need to get done and implement it. Start with that, and then you can expand from there instead of trying to drink from the fire hose and get all that information brought in at the same. You need to try to limit your scope.