DX Unified Infrastructure Management Reviews

The primary use of our UIM product is for monitoring Windows and Linux servers.

Provides a comprehensive monitoring solution for our open systems and also the mainframe. That is our biggest benefit. The comprehensive monitoring aspects of it with over 200 probes available and one robot.

The vast array of robots that are available.

We have a customized auditing feature that we have set up to ensure configurations are deployed to each server as per our desired model. We have 104 models and 104 packages. All of that has been custom written. If the product had something like that built in, I think that would be a benefit. 

That is an audit that goes to each server each night and verifies that the setup is according to our desired model for that particular server type.

We have a few customized routines we have had built for our scale. It is beyond auditing. We had to come up with a few alarm hubs and concentrated hubs for different segments which are not a standard use case. I would tend to think, in enterprise solutions, our use case would be sort of typical. So, we had to do some work to make what was more of a business class solution work at the enterprise level. 

UIM is very stable.

Scalability is good. We have had to do some customized horizontal scaling solutions that we fed back to CA, which we think they are incorporating, but it is scalable.

They are actually very good. Technical support is good. If we get to product support, the product manager is responsive. We have expressed some concerns and areas for improvement, and they were addressed.

We had no previous solution. It was the early days of Windows NT 3.5.

We decided to go with CA in the late 1990's, and we have been with CA ever since.

Initial setup was complex, given our business use case. We are not hosting standard back office solutions. We host a really complex set of solutions which we wrote. Another division handed us the opportunity to monitor their solutions as written, and UIM was very useful for that.

An opportunity with licensing that was presented back in the late 90's which gave us pretty much site license for any of our products. Then, once embedded in our operations and development, it was hard to dislodge. That was our primary reason. There are a lot of vendors out there. Everybody has a bell or whistle that is better than the next guy. It is what is integrated and what is your support kind of deal that you can do for me?

Most important criteria when selecting a vendor: integration and support. It has alway been well-integrated and the support is good.

• Cloud monitoring

• Easily skinnable
• SAP monitoring
• Reporting Tools
• Probe packages and probe deployment

Through action automation (using NAS and robot actions) we got reduced operational effort, improved operative tasks, and optimization of resources.

I would like to see improvements in the maintenance mode area. I would like to set only a probe from a robot or group in maintenance or place profiles from probes in maintenance. Making a GUI with criteria such as selection by robot/hub/probe etc.

I've been using it for five years.

No deployment issues encountered.

No stability issues encountered.

No scalability issues encountered.

We didn't use customer service.

10/10 CA Support team is excellent we get from them issue resolution within a single interaction and anchoring alternative solution.

Previously we had multiples tools to monitor IT infrastructure and other apps e.g ManageEngine to monitor databases, Nagios to monitor telco devices (switches, and routers, etc.) We switched to CA UIM because we had a lot of issues when we tried to correlate events and get unified metrics from our monitoring tools databases, in the same way we had maintain multiple tools and open tickets with multiples vendors.

Initial setup was straightforward. Installing and deploying software was very quick and easy. We have experience with similar tools implementation.

We did the full implementation in house.

Licensing model and packaging:

Server pack - Charged per server.
Server and application pack - Charged per server.
Service response time advanced pack - charged per site.
Ping pack - Charged per device.
Network advanced pack - Charged per device.
Flow analysis - Charged per device.
Storage pack - Charged per terabyte.
Ecometer pack - Charged per device.

We also looked at OpsView and SCOM.

Monitor governance and solution sizing are key topics to start solution implementation.

• CA UIM Solution is very easy and fast to use, it offers both agent-base and agent less monitoring options. UIM have lots of monitoring options that help us to manage our Enterprise heterogeneous monitoring demand.
• Database monitoring - helps us with monitoring critical database instances. We have many different databases from various vendors -- Oracle, SQL, Sybase -- and UIM helps us with managing all of them in one place.
• Infrastructure monitoring - Helps us to manage the data center, critical Ci's and v Center.
• Application monitoring probes - we are using Middleware application probes for monitoring critical apps.

It helps us to monitor critical servers, applications services, etc. We have many databases, hundreds of probes, and thousands of applications. This tool helps us manage all these and gives us insight into what's going on in our infrastructure.

There are several areas for improvement, among them are --

• Reporting capability - We're unable to drill down to the specific data that we'd like. We're also not able to get some data, such as trends, that we want when trying to determine root causes of issues in our applications. This is particularly frustrating because we have so many applications that we use and depend on that we need to know with some certainty what the issues are affecting them.
• End-to-end application monitoring probes function - We're quite large with over 500 probes, and there's no ability in UIM to monitor all of them through our many, many applications. We're relying too much on middleware applications to do this.

There was no issue with the deployment.

There has been no issue with the stability.

There have been no issues scaling it.

8/10

I've used them many times, and they were OK.

It's very easy to implement and use. If you want to onboard devices and applications, it's just drag-and-drop. Very user friendly.

In-house.

It has helped us cut manual efforts and reduced manpower. It's helped us with automation. ROI 50%.

This is very good product for monitoring as it has lots of functionalities and probes.

The most valuable feature for me particularly within the enterprise space and what CA does is that they provide an integrated central approach to monitoring. No matter what tool we're using whether it's a phone system or whether it's a storage system, whether it's Unix or Windows, whether it's an access point or a firewall, it can all be seen in a common pane of glass where you can see everything. All of our staff can all be trained on the same tool and then other people can see all the dependent infrastructure within the same tool.

It allows our communication to happen much more effectively. The primary benefit for us at this stage in our approach is that we've been able to cut outages by about 300%. For us in terms of an integrated management approach, it's an interesting question because the benefit that we get, it's two-fold. One we're monitoring everything so we're able to proactively see things that are going on but we also end up really driving the charge on technical debt. Because once we put everything in infrastructure monitoring, we have systems that maybe haven't been updated that now come front and center. If you have a storage array that you have been meaning to upgrade or you've been meaning to optimize, once you put it in monitoring, before we even start alarming and cutting outages, people tend to clean up those systems.

3 years.

You end up pairing off a lot of technical debts so most of that 300% improvement, I mean reduction in outages actually didn't come because of our monitoring, it's because we put the monitoring in place. That's the hard work that's done by our infrastructure teams in order to support it. The solution is very stable. When you're in the outage management business, the last thing that you want is your own solution to be going down. It doesn't speak well for the whole solution but we get a tremendous amount of stability from the system. We have redundant nodes and all that stuff all in place. The last thing we usually think about is the system itself going down just because, to be honest, in the five years that we've been running it, we haven't had any issues.

It scales fairly well. I mean the key with scalability and I think this is true with anything, you have to be able to configure it. The challenges, I think some people will take on an enterprise piece of software and not bring on the people to get it tuned and configured right and then it consumes a whole bunch of data. Then there's this fear of like, "This product isn't scalable." It's like, "Well, you have to tune it to takeout all the default data that you don't want and things that aren't helpful. A lot of enterprise solutions are going to give you the whole kit and caboodle right when you install it and you have to know when to turn it off. I think people end up thinking that something isn't scalable only because when they take it to the enterprise level without tuning it, it doesn't.

We had, to be honest, some of those pains ourselves, we were generating way too much data in our reports, we were generating a ton of network traffic. Once we tuned it and figured how to get just what we wanted and get just the right amount of metrics, our storage numbers and all that came down very reasonably. Serious technical support gets the job done. I wouldn't consider it to be the strength of the product. I think there's a lot of really smart people who run CA. Their product is in other solutions, their products are being implemented by an army of consultants. CA is largely implemented, well from my own experience, not by the consulting team but by hands on guys who work there which means they know the product. I don't think there's as much dependence on the support.

You start with the product. I will look at the product, how you it’s managed, how it’s run, how it’s designed. I’m an architect by trade so and that’s what I did my graduate work in. I'll look and see whether they have an architecture that’s scalable and supportable that they’re modernizing, they are making investments to make sure that it stays current whether it’s with programming languages or new tools or new modules. I look for a strong community base so I want to see something where people are working together outside of their confines of the company to develop either solutions, plug-ins or modules or to share information.

We've had some problems, some challenges with silos of support in situations where we would spend weeks working on something and then I'd finally escalate it and a senior technical person would solve in 10 minutes. The weakest part of the products we use is in the mobile space. We can't seem to get the products to work at all and we can't seem to find the mobile team to even talk to about it. Their mobile app is Circuit 2003 you know kind of very crude. It doesn't have any interface, it doesn't actually connect to our devices. I would say the strongest area for them is going to be in mobile. When you're running like we are in operation sweep, sometimes my guys are only one deep and they got to use the restroom and I send them in there with a mobile app in case something goes down but I can't do it if the mobile app doesn't work.

Generally when a company becomes serious about operations, older companies have been around, they understand the value of operations. Younger companies or some people might say more entrepreneurial companies don't have that strong operational focus. I think one of the cornerstones for that is when you get your house in order in terms of operations and infrastructure so that you're not your own worst enemy. You're not designing solutions that don't work whether it's in the application space, you're not buying infrastructure that isn't performing or you're not able to be accountable with the people that work for you to get them on the phone or addressing issues because they simply don’t know about them.

Today’s enterprise software can’t just be about the product. It has to be about the kind of ecosystem that that goes into it and companies need. If they don’t make a choice to invest in that ecosystem, they miss out on some of the best resources in terms of staff and technical development. I’m looking for a company that’s going to be straight forward. If they can make money because they have a good product people will pay for it. If they have to trick their customers into buying the product by selling them one thing and then forcing them to buy another because they don’t want to lose the investment they already made, that’s not somebody I want to partner with long term. I have personally worked with IBM’s product, HP’s product and CA’s product. They all have their strengths and their weaknesses. I wasn’t actually a part of this decision to make the investment in CA but CA has stood up against all of those vendors.

Say, they as were strong when you got down into it, they have some areas they are stronger other areas, they aren’t as strong. We’re a university so we’re not really in the making money business. Our decision was really about just ensuring that our students and our faculty and our staff had the best experience they can when they interact with technology and that means it’s up. That means that it’s working, that it’s fast when it should be fast, it’s stable and that really drove our decision. It’s that universities are very unique environments. The original campus, in the network sense, was a college campus. A lot of the technology that’s used in enterprise was developed on college campuses. There’s a lot of driving factors in there that come on the operational side to try and keep it stable. We realized that in order to be effective, we need to be able to see everything.

There can’t be a fog and it says, “We know how the servers are running or We know how the servers in storage are running." We don’t know what’s happening with the network or we know what’s happening with the network but we don’t know what’s happening with the VOIP or the phone systems. All of this technology is dependent on each other when things are down. Ends up resulting in a lot of finger pointing. When you have a centralized integrated approach and everybody can see everybody else’s stuff, you cut through all of that. If you have tuned the products and you can pinpoint exactly to the problem, then everyone can be working together on a resolution as opposed to spending all their time finger pointing, “I think it’s your network. No, it’s your storage array. No. It’s your server.” That all goes away.

I think the more complicated space is the ones where you have a mature monitoring solution but it’s not a central integrated monitoring solution and you have to figure out how to trade, things you’re going to have to give up for that silo in order to move it into an integrated approach. That’s always harder that just taking something that’s had no monitoring and bringing it up from scratch. I would put them in the top of the eight, solid nine category. I think they still have some work to do integrating some of their acquisitions. Like I said, the mobile and support space I think. My team has been very effective. These are a lot of people that I don’t interact with on a daily basis.

I think most of the time when people are looking at these other low cost solutions, they are not thinking about what they’re really going to spend. They’re not thinking about whether it’s sustainable, the solution that they’re putting in and they’re really going to see their return on investment. We had very specific needs in higher education that drove us towards CA. I don’t know if those needs would have been similar for instance when I was working in a marketing space. We’re watching a lot of websites and a lot of that kind of traffic. You need to look and see which product meets your particular space the best.

In our organisation we use this product mainly for server monitoring. OS, Database, URL, schedules, hub configuration and the dashboard and performance reports.

Management of servers, network and different types of probes and customisation of probes.

This is one of the main products for monitoring in our organisation,.e.g. our network monitoring sometimes doesn't capture interface traffic, during that time we use Nimsoft's interface traffic probe to check the interface utilization.

When ever there is an IP address change to the server and even changing the IP inside the robot.*.cfg file it won't update on IM automatically, and we have manually validated the probes. This should be improved.

I've used it for the past few years.

During deployment, not many issues.

The robots are a bit unstable sometimes, we often encounter robot inactive alerts even though the service is running. We have to restart to fix this issue.

Scalability in terms of how vast you can monitor, Nimsoft is a very useful product.

Customer service is quite okay, however whenever we face issue with probes, they only suggest to upgrade the probe to latest version and check if the issue reoccurs, and if it reoccurs then they troubleshoot, this should be improved.

Technical support provided by CA is quite good an they have a very useful knowledge base. Most of the solutions we find are on the website.

Nimsoft setup is quite straight forward, however, later stage usage and customisation depends on the requirements. A few probes are a bit complex, but the basic monitoring probes are easy to configure.

As mentioned, the initial setup was done by vendor and their level of expertise is good.

This product is very useful compared to other products i worked on like MS-SCOM, CA NSM, an IBM Tivoli.

As i have worked on this product for quite some time I personally think this product is the best in terms of overall infrastructure monitoring.

Server and database availability as well as performance monitoring in my environment. We also monitor URLs using UIM.

We now use only one console to monitor all servers, the database and the network. It has allowed us to consolidate four previous tools (BMC, HP, SolarWinds, Nagios).

It is easy to implement but requires good planning. If you don't plan the number of probes it can lead to an event flood. It is also reliable when it comes to monitoring.

Reporting capability can be improved especially when it comes to availability. Additionally, automatic baselines can help reduce the count of alerts.

Currently, we are building a visibility dashboard which allows our level-1 and level-2 engineering teams to actually look at the events, and be able to designate which ones are critical, which are important. It alerts them. They are able to simply look at a dashboard to be able to predict and resolve issues.

The CA tools allow to me to get into detailed transactions for custom ranking, and be able to make predictions. It also gathers data. Some other tools may be good at one area, but not good overall, including the mainframe.

CA has been very famous for the mainframe. I think that's actually the edge, by the way, for CA products, as there are many, many dotcom products that don't touch mainframes. I think that CA should still keep that. 

The area where CA is changing is, I'll give you an example. The fresh technology should have been out a long time ago. But it appears to be we're catching up on the UIM, because Microsoft basically says they are not going to support the fresh, starting next year. That means they are a little bit behind on the waves in the architecture, etc. I think that can be improved.

Another thinking is, if the developer can get to the level-1 engineer level, they will develop a better product. I suggest the developers at CA really work as a level-1 engineer, because then you have a different sort.

Infrastructure monitoring.

It preforms well, I mean it scales well. Handles about 17,000 servers. So it does pretty well. 

• Basic infrastructure
• Moving into cloud monitoring in UIM
• Resilient
• Pretty stable

Also, I think it's the architecture, the way the hubs are designed, the way that it scales, that it can be grown. That's valuable, in a large enterprise.

The large library of functionality; not having to go to multiple products to monitor different things.

Being able to report on the monitoring configurations and find out where you differentiate from standards. If I've deployed 500 probes to monitor Oracle, and I want to know that they're all monitored the same, I have no way to do that now.

Stability's been good.

The biggest problem is certain what they call "hubs." Different releases of different probes can be problematic, to get the right versions to work together. Or to find out if they scale or if they don't. So you've got to do some testing.

Scalability is dependent on the probe that you're using. Some probes scale really well, some things don't scale really well. So monitoring VMware may not scale as well as monitoring a cloud architecture. You have to test what you're doing.

I haven't used tech support for UIM recently but I have used CA's technical support in general. I would say they're responsive but can take a little time, if it has to go back to development for a review.

It's pretty straightforward.

When investing in a vendor, what's important to me are 

• software quality
• responsiveness
• communication.

I rate it an eight out of 10 and that's only because I think it can be better. I think as a competitive products in the market, UIM is really solid. A few changes could make it better.

Make sure there are staff to administrate it, after it gets deployed. And ensure that after CA delivers, that you have the ability to follow through with the rest of the implementation.