What is most valuable?
We really like the dashboards and their latest release. By going to v10 it made our lives a whole lot easier because prior to v10, we had to put all of these assets, these Riverbed arguments, basically JVM start up itself. Then when they send out a new release or a new patch on it, the library changed to 64-bit. If that library changed, we have to go to all the different JVMs we have implemented and we would have to go in there and change all of the start-up arguments. When we go with system wide enabling, we don't have to mess around with the JVM arguments anymore. We just control on the backend of AppInternals what we want to instrument and what we don't want to instrument. That makes it a lot easier.
What needs improvement?
One thing that I can't do with the AppInternals - we have put in a request as an enhancement that we would like to search for items. If we're searching for a URL and we want to know was it a get or was it a post.
Those are attributes that are of interest to us because we can have a static URL but rather than just seeing the performance time, the metrics and everything around that, we needed a way to go in and carve those out. To find out which one was a get and which one was a post. We would like to see those columns. We would like to have those columns available that the customized columns had.
With v10, Riverbed has done a very good job and we've been using it a lot. There's actually more features in v10 that we really haven't gotten into yet. I'm planning on making some more time to dig into those.
What do I think about the stability of the solution?
In terms of stability we haven't had any issues there. There was one time when we were trying out different versions and when we tried v9, for some reason it didn't working well. We then actually would go ahead and hang up our transactions, where they wouldn't be able to go through the application.
That was something that they corrected. With the later part of version nine, we went ahead and rolled it out and the problem was corrected. We did see that little bump in the road where we caused a little bit of aggravation. We always try things out in our non-prod environment first to see how they are going to work. The good thing is we didn't have it in production yet. We were able to catch this anomaly, happening in the non-product. Riverbed engaged, they were are extremely responsive and they got that corrected for us. Apart from that one little bump in the road there, you know that we've been happy with it. We have been happy with it and it's been pretty stable. It has not caused any issues.
What do I think about the scalability of the solution?
This team here is definitely scalable. Right now we have about 100 licenses in prod and 100 licenses in non-prod and we are not using all of them yet, but we are getting pretty close. In terms of scalable within the organization, we do have other groups that can start using them more. Right now, I think we are mature enough with this to where we can get a lot of our developers to just get hands on to where they can start learning it.
For the most part, it is the middleware team that uses this tool quite often, as well as other stuff in the low team where they use LoadRunner. In terms of troubleshooting, essentially it is the middleware team who is engaged with making sure that we have the Linux guest and that the applications are deployed, and they're all surfacey. That's what the middleware team is tasked with. In addition to that, the middleware team knows the product so well that they know how to troubleshoot it.
We go in there and we use it a lot, but one of our goals is to go back and try to engage some more of our application developer folks. Never too early, they could get more of a hands on.
How are customer service and technical support?
Those guys are extremely helpful. If I were to rate them, I would probably give them a 9/10. The reason why I wouldn't give them a ten, because we had on a few occasions on a Friday, where we tried to reach individuals there in customer service and they weren't available. Even when we got them a message, the intake person said that we will contact somebody but they didn't. Then that Monday came around and they still didn't call us, so on Tuesday we called them and we said, "Hey, well, what happened? We had some issues and we let you guys know about it on Friday afternoon, and the on-call person wasn't called and nobody called us back." It wasn't a critical issue, but it was hot enough to where it would have been nice to get some guidance from Riverbed, but they went ahead and let us down, but that was only one occasion. That's why I wouldn't give them a ten, but they would definitely be at a very strong nine because during the day, like Monday through Friday during their core hours, they're always available, extremely responsive. Then they always get back to us and they also check in.
Let's say we're working on an issue. Periodically they'll reach back to us and say, "Hey, is everything okay? How is your testing going? Can we help you out with anything else?" They're very good on follow-up. They're very good on follow-up and they're extremely responsive.
Which solution did I use previously and why did I switch?
We used to have another tool prior to Riverbed, which was CA Wiley but it was just not as robust. At one point we had found out that that tool had a memory leak in it which was causing our applications to go a little sideways. It took us three times of climbing the ladder to purchase Riverbed, and that was just convincing the executives that this was the way to go. Everybody likes it.
Which other solutions did I evaluate?
We kind of got hung on Riverbed. They came in and they did a demonstration for us, and we were so delighted with the demonstration that we kind of grabbed onto it, even though it was just a POC. We previously used another tool, we just didn't like it. It could not compare to Riverbed with the ease of use of the Riverbed suite. When we went to the Riverbed suite, we really latched onto it. It was very easy to set up. We really used it a lot, but then when the POC ended and it came to the point of making an executive decision on if we were going to purchase it or not, it went all the way to our executives and it was turned down. We weren't able to purchase it because apparently we had a long contract with CA, and the powers to be said, "Oh, continue to use FCIO because we won't have to pay for it."