DX Performance Management Reviews

• Network performance reporting 
• Incident monitoring

The network teams have transitioned to becoming proactive in responding to network performance issues, rather than being just reactive to outages.

CA PM is the single pane of glass, providing consolidated views for all of the CA tool sets, for viewing all reports.

CA PM Business Hours Filtering: I understand that all monitoring systems have defects. The Business Hours Filtering does not always function properly. Sometimes, when applying business hours to CA PM reports that are 30 or more days in duration, the report generation times out and does not display results. 

We have other CA PM reports that, when we apply Business Hours Filtering, the report results displayed are the same as without the filter. We are not sure if this is a defect in CA PM or if it is a result of our complex configuration (folder structure) and application of business hours.

We would also like to be capable of reporting network performance with a report strictly focused on the times outside business hours, which CA PM does not currently support. We have discontinued the use of the Business Hours Filtering until CA engineers are able to resolve or offer guidance.

Initially, we had under-sized the ADA and NFA.

No issues with scalability.

Good.

NetQoS was acquired by CA.

Complex. CA PM can be complex to build and configure. Creating the folders / groups / sites required establishing many rule sets.

I have no experiencing with pricing.

Netcool, Solarwinds.

Instrumentation could take some time, depending how much reporting customization you plan to employ.

The most valuable features to date are the alerts, reports, and integration with other CA products:

• The alerting feature allows us to quickly identify and begin remediation on network devices that are experiencing performance/capacity issues.
• The reporting feature has allowed us to quickly share information with customers, keeping them up to date on critical devices that are being monitored.
• The integration aspect is also important, as it allows us to feed inventory to the system from tools designed to walk our networks, and also pull in flow data for additional troubleshooting in one place.

Utilizing the capacity/performance data alongside the flow we are able to quickly determine the root cause of impact. This allows us to reduce mean time to resolution on network impacting events due to capacity/performance.

There is always room for improvement, which CA continues to do based upon community feedback.

One of the greatest challenges we have encountered are the polling cycles. We currently poll at 5-minute intervals, but at times need to have up-to-the-minute data. Products we have leveraged from other vendors allow us at the click of a button to turn on up-to-the-minute polling for up to an hour. CA Performance Management does not have a similar feature. It may be emulated, however, by setting up a new group for one-minute polling cycles; however, leaving this rate on will impact data storage, so only administrators are provided this feature in our organization.

The solution has been in place for 15 months, and has already provided information helpful in network cleanup.

To date, we have not encountered any stability issues.

To date, we have not encountered any scalability issues.

I have not been required to work directly with technical support. However, based on feedback from the team that does, support often starts with the standard runbook, even after we have described similar troubleshooting steps to them. This results in lost time in resolving issues. Once we have moved beyond the basics, support has been handled well.

Our prior solution was dated and in need of refresh. We switched to the CA solution after going through a proof of concept and rating it against two other competitors.

The initial setup was mostly straightforward, with assistance from onsite technical resources from CA. The majority of issues we encountered were with our corporate requirements on server configurations and not the tools themselves.

Regarding pricing and licensing, keep in mind that bundling will provide the best value. As a result, doing your best to anticipate growth and future needs may save you money in the long run.

We did evaluate two competing options from HPE and Alcatel Lucent.

Take advantage of the professional services and leverage your solution experts from CA. They will help you define the right solution keeping in mind current utilization and future growth.

The tool monitors device performance, including performance utilization, bandwidth usage, configurations, and the status of CPUs and memory.

The tool helps us understand network performance. 

DX Performance Management should increase high availability. If one of my two data centers goes offline and the other is operational, I need a seamless transition from the primary to the secondary data center. It's crucial to maintain consistent monitoring of the devices during this transition. Therefore, I require continuous, 24/7 monitoring of my system using this tool.

I have been using the product for five years. 

DX Performance Management is stable. 

The solution is scalable. My company has 500 users. 

DX Performance Management has support delays. 

The product's deployment is easy for Linux admins. 

The tool's pricing is based on the number of network devices. 

I rate the product an eight out of ten. 

We are gathering the SNMP data from many input devices, especially those which are used to monitor the status of the switches, the routers. It is very helpful to know about any CPU spike, rate of the CPU memory utilization of the network devices, along with interface utilization.

Also, we are using this product to get some QoS metrics from the network devices within our network. We are also using the Network Flow Analysis, to analyze the traffic, the conversation between the offices.

We have one more product called SNMP Trap Explorer which we are using to receive the traps from the network devices. The network devices are configured to send these traps to the CA application, which is all set on our internal network. Then we use the traps to send out the notifications to our internal support teams.

It is very easy to add devices; just be aware that it requires SNMP to be enabled.

It is very easy to use because most of the administrative tasks through the UI are documented. If I have any doubt, I can just go to the CS support. In the UI itself, you have a lot of helpful information on each administrative task, such as adding a special profile, conferring the vendors' specifications.

The documentation is very good, they have the documents up to date.

It would be helpful if CA provided online training for its customers. Whenever a release happens, the latest version, we have meetings with CA meeting and they explain the latest features and enhancements, which is really helpful. But then I'm training my internal support on our tool, if there were any training that CA provided directly, online, that would be good.

It's very much a stable product, because we are running it stand-alone. We have a two-tier architecture we run on. It's a stable product.

Currently we're not facing any issue with our infrastructure. There are around 5000 devices monitored using CA Performance Management. To be frank, I don't have any idea what the capacity of the product is, going forward. But currently, I don't see any issue with the capacity, the amount of devices. 

The 5000 devices are monitored from data collectors housed in three datacenters.

Currently we're on version 3.2; we had some issues while upgrading to 3.2, but we received very good support from CA within a short time. They were able to resolve the issues and communicate to us without any delay. We are getting a pretty good amount of good support from CA.

Normally we submit a case using CA portal, their support website. They'll reach out to us within 24 hours after we submit it. If we need support from the local, India CA support team, they are very flexible and can transfer the case, based on the customer's time zone. Because I'm based out of India, I have some issues with working during US time. They will help me to work with one of the engineers who is based out of India, I think. That kind of flexibility is really very good. 

Also, documents are very available from support.

I wasn't involved in any decision on that. But it's all based on management's decision.

It took some time because we could not just upgrade production. We have CA PM on two environments, one is Development, and then we are running on Production. Whatever changes or upgrades need to be made, first we'll perform then on Development. Based on the results, if anything is found, or if the activity goes well, then we'll proceed with the Production system update.

There is a really good amount of documentation, and whenever we need help, they will set up a WebEx session directly, and they'll help us during the implementation, installation, or upgrade activity. 

If I were to talk to others who are considering this product, I would explain the features, that it's easy to use, and its scalability, stability, and about the capacity it has.

I would also tell them that CA is upgrading the versions to stay current in the market, that is done very frequently. Whenever there are changes made - take the example of a Cisco router - if there is a new series of router, CA will include that in the next release.

Also, there is a good amount of vendor certification which comes with the product. That's all factory-loaded, no need to load any custom-made files. Most of the metrics are calibrated and captured from the devices based on the defaults available from vendor certification. 

Considering all these factors, it's really a good and easy product for analyzing your network performance, health status, and the quality of services.

I rate it a nine out of 10, only because I still need time to explore it more. It has a lot of good features, but I am still in the learning phase.

• Multi-tenant platform
• User-configurable roles & permissions with granularity
• Dashboard-oriented
• Easy enough to customize

Mainly as an MSP configuration, the multi-tenant model helps us to segment the customer and users, as well as providing our operation the capability to oversee all tenants.

Today, tenants are bound to a data collector which requires a lot of networking and resources to configure.

In the context of the multi-service provider and the need to have customer segmentation either for IP overlapping or Security (data protection), multi-tenancy is required. However, in order to do this, the tenant (IP Domain) need to be attach to one Data Collector by Tenant. In the MSP context, it is equal to 1 DC per customer and even for small customer (couple device) this is not too scalable or rather expensive.

I have used the platform since the beginning and in production for at least two years.

Our multi-customer configuration is challenging, as well as to have redundancy over two data centers.

We encountered database issues due to the IOS speed requirement of HPE Vertica, as we are on a VMware environment.

Redundancy over two data centers is one of our challenges, as well as the fact that we use a lot of data collection, even using Docker containers for it.

Support is OK for the standard use of the product but as our configuration is complex, they often have to refer to engineering.

We were previously using eHealth and APG.

As we started from the beginning, it has improved a lot since; once again, our solution remains complex due to our usage.

Implementation was with a vendor team; at the beginning, with CA Service, and then with CA engineering, as we do DevOps.

Before choosing we also evaluated SevONE, APG, and InfoVista.

This is coming along; great progress. The "3 series" looks promising.

The most valuable feature to us is that it provides us with NetFlow data, which is essential to our business.

It allows us to see historical data that shows what type of traffic is going across the circuit. When we have capacity issues, we can isolate those issue and we can block when we get high traffic.

There are some issues. We started supporting UTA and began to see both major and minor problems. We're able to get support for these problems, though.

We've had no issues with deployment since implementation.

It seems stable. We're using NetQoS.

This is the scalable option, and it's why we need it and have it.

Technical support is very responsive and knowledgeable. We get tier-one support.

We had InfoVista when NetFlow was an emerging technology. CA really stood out and that’s why we brought them in.

Initial setup was pretty complex, but we were familiar with it and got help from CA.

Not too many companies would implement this. If you're a large organization, make sure you staff correctly. It’s a large application and you need a group of people who know Linux and NetFlow. Have the right skill sets.

We use it for monitoring up/down, fault management, and trending.

Although it's a really good tool, it very much needs to be rebuilt so that it can be updated on a regular basis.

One of the ways this product has improved things would be disk space. Since it is monitoring a lot of our infrastructure, I can get it to run a report showing, for instance, what root drives are in the critical range in terms of being full, like 90 percent full, and disseminate that information to the other areas of the organization. I can let them know that there is not a problem now, but there will be in the near future and that if you run out of drive on your root drive for your OS, bad things happen. It's not a pretty picture, when that kind of thing happens, to recover from. Running that report showing which servers and file systems are almost to the critical level is a good, proactive use.

Also, the ability to very quickly and graphically navigate around, being able to zoom in to a timeframe, apply it to all the other panels - things of that nature - are all very intuitive. When I give someone access to the system, the learning curve is pretty short and they very quickly start to discover how easily, with a few clicks, they can get to what they're looking at and understand how the system works with minimal effort, unlike some other systems where you would need some training to know what you're looking at.

The metrics that it's able to track and that it's able to trend,

Keep in mind, because we have not updated, it's to the point where I've got to build a complete separate system with the latest version, and then map over all the users and customization, etc, then swap IPs to bring the previous system down so that we'll be on the latest. There are about two years worth of updates that I'm missing out on at the moment. I'm hoping that there are a lot of updates and nice, new features that have been added. So, really, I've been in the dark ages as far as running for a while now, but the feature that is inherent to its core, of being able to graphically represent a trend and status, is nice.

This may be available by now but for server space, when it comes to the disk file system, I'd like to see that graphically represented, or the trend, rather than what I'm seeing now. I don't know what the latest version has, all the nice, new features, until I get that project underway.

It's fairly stable. This still falls under the issue that the current version I'm running is two years out of date and is less stable, I'm sure, than the very latest version. But stability has been good.

One of the best features about it is that it's a ground-up application from CA. So, there are typically more features added more quickly and there are more updates to the product on a regular basis, unlike some other products.

We have never had a problem with scalability. We've always been quite well under what it can do.

Tech support has always been good. A lot of the people that I've known for many years working on the Spectrum side of things have moved into the Performance Management side of things. So tech support has been good.

The response time is good, and typically, with any case with CA, I am sent surveys that I fill out on a regular basis to evaluate performance, how it was handled.

They will come by and visit occasionally, and they're really good about being available when I need help with looking at new systems, to show us what they have to offer.

We did not have anything previously, except maybe something along the lines of the free MRTG. But previous to that, we were on a Unix-based system. The biggest thing is trying to tie them all into a centralized database, or a CMDB, which is a project that we're still working on, making progress on, so that everything feeds from a centralized database and is aware of configuration items. That way we can interconnect all these applications so that we can add automation and things of that nature.

It was a bit complex to set up. I don't know how much that has changed. But it was fairly complex to set up.

I didn't get any training or have assistance setting up and it did take a while to do. I followed the directions. It was straightforward but there were a lot of steps to set things up at the time. Keep in mind, again, it was quite a long time ago that we set the system up.

I believe the product's pricing is a good value at this point. We're currently trying to internally evaluate everything that we use. It's moot in that it falls under the same umbrella that Spectrum does, and Spectrum is an integral part of the tool. It is used by a lot of departments within our agency to monitor the network up/down. So, it's a part of that umbrella, they're included together. With that, the pricing is fine.

Many years back they changed the way that they did licensing and that was good. Since then, everything has been just fine.

I would give Performance Management an eight out of 10, and I might even move that to a nine after I've had a chance to upgrade it to the two or more years' worth of features that have probably been added.

It's quite usable. It's quite good to see something graphically, it's a very graphical application.

The CA Performance Management solution wasn't suitable for our small company and it was not easy to install. The requirements you need to install it are really high and many customers are complaining about it. After using it for a while, we've decided not to continue using it and moved to a different solution.

One feature I like about CA Performance Management is the certification of the devices. This solution is very functional, but it's not suitable for medium and small customers, which is the case in most of the markets in Europe. There are a few huge customers, but most of them are taken directly by the provider.

It is a very good tool but I don't think it is suitable for the European market right now. Huge customers might benefit from using it as most of their customers are directly managed by the vendor and not by partners or freelancers. An improvement would be to reduce the footprint of the installation of the solution. I believe there is room for improvement. 

It is a very stable solution and it works, but the initial footprint it really high. And because you can add extra performance and more servers, you can escalate very quickly. Therefore the initial installation and footprint are huge.

This solution is not very scalable. For example, a case manager in another company was able to handle a thousand devices with only one server. If the same customer with the other company logs into CA Performance Manager and needs to do the same, he will need seven servers. How can you explain that? Even if the customer has free access, it's not possible. It's not reasonable.
The investment you have to put into strong enough software is high. So we've lost many customers because they didn't like the replacement even though it was free for them.

The technical support from CA Performance Management was very good. 

The initial setup was really complex. You can probably set it up online in one week and it will start working and collect data right away. So that is one good thing I can say about the performance. Most of the devices that you find in companies and enterprises can be incorporated in the database of the product so that can be monitored correctly. 

CA Performance Management is a solution with different models but the requirements for the basics that you need to install it, are huge. This is not a good solution for small or medium-sized customers. I rate this product a six out of ten.