Broadcom DX Application Performance Management Reviews

We are using APM for troubleshooting and because of this product, the delay to solve and to fix issues has been reduced by 40% on average. We also use the APM data to analyze and predict if an application will work.

For us, the application is a new eye on the same data. Its quality and the way we access it is going to make the difference between an application you keep using and an application you won't. There are a lot of DevOps methodologies to improve the time to market and to deploy applications quicker. Sometimes it's not enough because at the end, the application expands some issues and the isolation of a root cause is quite complicated. We analyze about 3,000 incidents and predictions, classify them, choose our toolbox, including APM tools (the CA APM solution), and we are able to predict, to anticipate incidents and to assess a risk for an application to go live.

An application is quite complicated in the environment of a software reliability engineer, because our applications are like a black box. Thanks to CA APM we are able to transform this black box to a gray box by using the transaction trace functionality services. For me it's the most valuable service of the solution.

The product displays some graphics and sometimes we want to have some different metrics in the same graphic but it doesn't display in the same scale, so it's quite complicated to see the metrics. If the product would be able to manage a double scale, it would be perfect.

We measure the limit of a solution by the number of users. The application is developed in Java and only 10 to 15 users can connect to the application. It's quite a limit for massive deployment in a huge company like Orange.

Our scalability is great because you can add servers. It works.

I don't use support, I do everything myself.

It was the only tool in 2006 to provide the transaction trace feature. As mentioned before, it transforms an application from a black box to a gray box, that's why we knew that we needed Introscope (now CA APM).

It was quite easy. The solution is easy to deploy and we had the support of a vendor team.

We deployed it on 5 or 6 applications and now it's deployed for over 200 applications.

Detailed alert settings give more control on when they should or should not appear.

The tree-structure tracing shows where we are spending the most time in a call-by-call sequence. But the tracing capabilities of other APM tools are more advanced than CA.

APM tools are essential for finding performance bottlenecks and pervasive issues. It gives a high level point of view on where there is a problem. This tool, in particular, is not as good at deep dive analysis. Newer versions, like 10.5 with AXA, Dynatrace, and AppDynamics may be better at going deeper into root-cause analysis.

Deep dive analysis and performance need improvement. Introscope scalability performance lags behind for our team, which looks after 3,000 JVMs and millions of metrics.

I have used this product for a half a year.

The product is stable most of the time. However, like any software out there, there are bugs. Few crashes have been experienced, but incorrect metrics and improper collector behaviours have been observed.

We have had scalability issues. Performance slows down with the number of JVMs our team supports.

I would give technical support a rating of 7/10. Few people are available to support the product. However, good, proper attention is given to urgent matters.

We didn’t have a previous solution.

I was not involved in the initial setup.

This is out of my scope.

Other alternatives are being actively considered.

I like the Transaction Trace Viewer and stalls feature on the front end/back end within the investigator.

It has helped to stabilize the environments and also quickly identify issues some while they are occurring and some proactively.

The extensive monitoring alert gave the ability to act quickly before an outage happens.

We haven't implemented the latest released version yet but I do believe there's a lot of improvement needed.

There's a lot of room for improvement in administration which I believe CA APM is slowly getting there, with their Agent Command Center. However, the interface and management of the entire APM infrastructure should exist via GUI.

I have used this solution for five years.

We have scaled CA APM a lot over the last few years. As time goes by, there is an increased demand for monitoring leads to underperforming enterprise managers and/or other memory issues. We are running on VM. However, CA recommends to run on a physical machine with lot of compute resources, which wouldn't be applicable to all companies with their standard environments and also in terms of the budget/funding that it requires.

The customer service is good but not excellent. I would rate it a 7 out of 10.

I personally didn't use any other solution, but my company did use the IBM APM tool.

The setup is complex. It takes some reading and understanding of the architecture along with some prior skills.

We have implemented it in-house. My advice is to install a PoC environment and familiarize yourself with all the components.

Compared to other competitors, this product is a little flexible with its pricing/licensing and is somewhat less than the other APMs in the market, with negotiations.

My advice is to implement the product and leverage all of the add-on free components. You should also compare it to other APMs in the market.

In many organizations I've seen this product working in, SEV1s have been reduced quite a bit. One particular organization, within the first 12 months, reduced their SEV1s by 90% just by introducing this product. They had previously been using a competing product, so it really speaks to the value provided by this one that they weren't getting out of the other products they were using.

Suddenly, we were able to evaluate based off how code was being traced. In one incident, it saved us 18 hours in finding the actual problem in the logic for that application once we had a problem.

The most valuable features I believe would actually be two things I would go into: The insight it gives into the applications that it's actually applied to, and the flexibility to do many things with those metrics, and also feed your own metrics from external sources.

I think as we're all moving forward to automated deployments, it'd be nice to have that out-of-the-box with this product.

Also, scaling it, the data nodes writing automatically would be a very nice feature.

We have had large environments, many millions of metrics feeding in, and never have we had to dedicate a single resource to maintaining it.

Actually coming in once in a while for maintenance, we do have to put some work into it once in a while, but there is nothing to the magnitude of having to have somebody dedicated to it.

It is scalable. We can keep adding collectors that store the data, as needed. We haven't had any issues with that so far.

Technical support has always been very happy to help us. They offer a community, which is often very helpful, where one can go to find their own answers if they'd like.

We already had a competing solution. We weren't getting the value we needed out of it. We were still having problems. We couldn't find the actual root cause very easily with that solution, so we started evaluating others.

We looked at many others and this seemed to be the best fit for our organization.

Really, the most important criteria when selecting a vendor is value for dollar. As any business, we're looking for values and it's not just, "Is it better?" But, "Is it better enough to be worth the cost?" Or, if it's cheaper, that's also great.

The initial setup is actually pretty straightforward. The install is very easy. I don't think one would even need a document for it. It's pretty straightforward and the questions it's asking are kind of self-explanatory.

My rating just reflects some of the features we'd like to see. I don't think I'd ever give anybody a perfect rating. There's always room for improvement, but my rating is definitely higher than any other competing products that I've used, against it.

• Isolating issues, the dashboards, and the customization are the most valuable for us because they enable this tool to be used by multiple tiers. Let's say we don't know what the problem is when one occurs, we just know that we have multiple tickets generated, and there are customer groups down or experiencing slowness. These tools help us identify the issue quickly. That saves us a tremendous amount of money. That's the number one value.
• It is easy to use. That's a big one.
• The fact that it is a suite solution, rather than using just one technology.
• There are multiple offerings. That's nice as well.
• The agents are low-profile, and there is not a whole lot of overhead with the agents.

It sounds kind of corny, but it helps bring us together because when a problem strikes and it’s happening, there's the tendency to start finger-pointing at different groups. With this tool in the hands of a tier-one or a tier-two technical individual, or small group, they can identify the issues without needing to go back and forth between different departments, who are all saying "Oh, that's not our problem." It then turns out to be their problem in the end. That's the main benefit.

There are a lot of improvements to be done. One particular area is in the area of automation. That's a big thing for me.

• In my personal opinion, it's a bit of a weak point. I believe that they know that, and that they're working towards it. I would love to see that. I think they are heading down the right road. It’s a sound product.
• In APM, in triage, it's all about time and less effort. If an outage bridge or a severity breaks out, I want less people on the phone. I want a small group of individuals. I don't want every head of department on that bridge for hours on end. There are a whole lot of ways. That's a soft cost; but it's a cost: absolutely.
• I want to see more automation out of these products, especially in identity management. As we virtualize, and as we become more mobile, I want identity management to seamlessly move with the user and applications and systems. That's a hard nut to crack.

I have very high expectations for these tools. If it takes me more than a matter of minutes to identify a problem, it's not good enough. There's still a lot of growing to do. A lot of it is internal growth, too. It’s not just CA. It's not their fault; but our own company needs to adopt certain practices and change our ways. We need to be more open too, I guess, with more internal collaboration to use the tool to its full potential.

It is stable; and, more so, I like CA's direction. I am very inspired by it. It gives us hope.

The product has absolutely been scalable.

My team doesn't deal with technical support directly. We work with internal groups who are the actual customers of the product.

I was not involved in the decision process to purchase APM or in the setup. They have actually been in-house for quite some time, before my time; but I've learned to know them. I've learned their products. I've met with vendors, as well as other vendors and competitors. CA seemed to beat out everyone for our company.

Never be satisfied. Always ask for more, and always look for the next step. As long as the company's vision is really the future, then I think we're in good hands.

The most valuable feature of APM is gaining insight into the application performance so we can proactively take actions before the customer calls to report an issue.

We have used CEM to get insights into what the user is actually experiencing.

There needs to be more ability to report on the performance of the applications, so you could easily have a report run and give you all your performance stats for an application at the end of the month, or week, or after a release. This currently is not easily done, and you can’t schedule it.

In the beginning, there were issues operationalizing CA APM to work with our company. I wasn't involved at the time, though.

We have encountered issues with stability when best practices were not followed. It became more stable after using best practices.

We haven't had issues scaling it.

I'd rate customer service an 8 out of 10.

Technical support is an 8 out of 10.

I was not working in APM when we brought it in, so I don’t know for sure if it replaced anything.

I wasn't involved in the setup.

It was implemented with a vendor team.

We have an ELA with CA APM and it is very competitive with their competitors from that perspective.

Our company did evaluate another option, but I was not involved at that time.

Make sure to test as many application types as you can in your POC phase so you can work out issues early in your implementation. Also, make sure load tests are included in that testing.

• Proactive alerting is one of the key elements of APM and works really well. The ability to detect problems before they become widespread high severity incidents is critical when you are managing production systems. All of our APM is visible to our global support operations centres, and each service is represented by a RAG summary alert button. Any AMBER/RED status alerts are checked instantly by our 24/7 support teams.
  
• Ease of deployment
• Product stability

APM is the tool used by all support areas to monitor Java application and infrastructure performance. We have standardised support for the services running on the Java platform into one tool.

Dynamic instrumentation, at the moment, I think it is a bit clunky to use and it should be available outside the transaction tracer. It can only be enabled during a transaction trace session. Ideally it would be nicer if this could be done via the Investigator view, and the available options are shown with a right click. That way they could be done without having to invoke a transaction trace session.

I've used it since 2004.

No issues encountered.

No issues, it's one of the main strengths of the product.

No issues, the product has evolved to increase scalability on existing infrastructure which is very good.

Excellent, really top quality always.

Same as customer service, top quality.

No, but we have retired similar solutions in favour of APM, e.g. Tivoli.

Straightforward, but we had professional services from the vendor to guide us as well.

Combined, and the vendor team was top quality.

It's not easy to quantify as the product is used to prevent mission critical services from going down. We can’t really put a figure on it.

Two full time employees are needed to manage the whole operation, which covers all production environments and two test environments.

Not before, but we have evaluated others since, such as AppDynamics, Optier, Dynatrace and others.

• Stay close to the vendor guidelines at all time
• Don’t customise it too much
• Always patch
• Join the user community

We look at the load patterns and server health by looking at server health parameters. Heap memory is of prime importance as well as how the heap is behaving and how the memory is behaving under load, as well as the CPU utilization. These are generally the parameters we will check.

In a production environment we check all the parameters. We can detect any leakage, any problems, whether RAM is sufficient, and whether CPU is crossing a threshold. We can determine if we need to add new servers or RAM or adjust the heap settings.

It helps save resources for a company.

It also helps us look at business-critical transactions and how fast they are happening. It helps us pinpoint the methods or functions that are taking longer to operate and we can advise the development team to look into it and fine-tune those areas. They can make adjustments immediately.

For me, the most valuable feature is being able to check memory patterns. We check them when we do load testing or soak testing to see if there are any memory leaks.

A CA APM agent takes a lot of memory. That is one disadvantage. If you configure CA APM correctly it will still consume around 15 to 20 percent of memory.

We haven't had any issues with its scalability.

When you submit a support ticket it will take them two to three days to reply. They told us they would resolve issues within 24 hours, but so far that has not happened.

There have been times when we were doing performance testing, load testing, where it didn't capture the required information we were trying to monitor. We called the CA APM support team. They suggested we increase the RAM and restart the agents. That was the solution they gave us.

The setup is not straightforward. It is a bit complex. I have not configured CA APM myself, but I use it for performance testing. When we are doing performance testing, we configure CA APM for a pre-production environment which is similar to the production environment. We will do load testing and take all the parameters and observations and send them, along with our recommendations and conclusions, to development. But I have not really configured CA APM.

Dynatrace is the most expensive of these tools followed by AppDynamics which is "medium-expensive." Both CA APM vs Dynatrace and CA APM vs AppDynamics are a bit lower in price. Because of this, our clients insist on using CA APM.

In my opinion, if you can afford it, go for AppDynamics instead of CA APM. But if our customers don't have the budget, we'll go with CA APM.

We have a comparison chart showing the license costs for AppDynamics, Dynatrace, and CA APM and other monitoring tools as well. We provide that information to our clients along with the advantages and disadvantages of a particular tool. We provide our opinion on which it is best to go with but we also listen to our clients' opinions. We discuss this at length to finally come up with a decision on which tool to go with.

I use many tools, such as LoadRunner, Silk Performer, and IBM. I also use AppDynamics and Dynatrace. The last two are actually the best for me at the moment because they support even the database level. CA APM does not support that much. It will only give me minimal information about the DB.