A10 Networks Thunder ADC Review

Improves our performance and management, saving us money


What is our primary use case?

My primary use case is to use it as a software load balancer. 

Because of the industry that we operate in, we cannot use cloud. Therefore, we use it on-premise. We have 32 medium boxes, and that's what runs production.

How has it helped my organization?

A10 explained why the latency dropped significantly on a site that we have.

It gives me information on load balancing and offloading, which provides me information on performance.

What is most valuable?

I use SSL, TSL, and additional offloading. With additional offloading, this is where I can put my certificate on A10, as servers don't have the capability.

It has allowed us to smooth out our traffic.

What needs improvement?

The ease of use could have been created better. Some of the UI features are very primitive. Sometimes, wrong entries will go in and stay. I gave this feedback to the team who sold it to me. 

Also, some things like the traffic flow management take a lot of time to learn to use. While we have mastered using the feature, it doesn't tell us where we are going wrong or if something is breaking.

It will start failing if you have a containerized environment, which is why we have to start moving away from using the A10. If they could fix this, that would be good. 

It would be also be nice if someone could walk me through the solution’s Secure Service Mesh to optimize traffic within Kubernetes and containers, since I am not able to use A10 for rebalancing right now.

While I do use the security features on the load balancing, I find them to be primitive.

For how long have I used the solution?

I have been using it for two and a half years.

What do I think about the stability of the solution?

I can't complain about its stability.

My engineering managers, with help of the DevOps team, manage the deployment. They have become self-sufficient and, with zero impact, they patch the servers in broad daylight, meaning mid-week.

What do I think about the scalability of the solution?

I haven't needed to vertically scale beyond this box. It does scale horizontally.

The journey started with four servers and two websites and now I run close to 32 servers and service 36 applications, consisting of Web, application, and microservices. CPU utilization has moved marginally, by about three percent.

How are customer service and technical support?

Troubleshooting always results in us contacting the tech support team and have a solutions expert, who is part of the sales team, send us helpful information. The solution expert is phenomenal, which is not the same experience that I have with the technical support.

I haven't been impressed with the technical support, though. When we need help, there is less support staff than we would like in the time that we need them. E.g., it may take an hour and a half to get assistance during an emergency. 

I would like them to provide learning tips and a community forum. where users can share ideas. They need more detailed support articles on the A10 website.

If you previously used a different solution, which one did you use and why did you switch?

Before A10, I was using Network Load Balancing from Windows, which came pre-bundled and was primitive. I wanted to move to a professional version where I would have more control. I evaluated a couple of products and A10 seemed to be the clear winner. I had very simple parameters: the latency of response from our servers. The benchmark was NLB. When we compared the numbers with A10 and, when configured properly, it reduced the latency by roughly 15 percent.

There is absolutely no comparison between A10 and Windows load balancing.

How was the initial setup?

The initial setup wasn't complex. A10 helped us with the setup. 

It took seven days to go into production, which was due to number of sites that we were running.

Implementation strategy:

  1. It should have the ability, as a load balancer, to manage circuits.
  2. Not everyone should have access to it.

I created a replica of the production box. Then, I set up the service, which I put on the load balancer. For 15 minutes, I moved the traffic from my NLB through A10 and monitored the performance. Then, I moved it back and it gave me the confidence that it could run safely. Then, I did the same for all 36 sites that I run. Afterwards, I moved it over to my team.

What about the implementation team?

A10's solution expert helped us configure the box.

A third-party called Value Point helped us with the deployment. I was not impressed with their skills. 

For the deployment, there were two or three people from my team, two or three people coming from Value Point, and two from A10.

What was our ROI?

It's a phenomenal investment that the company has made.  

It improves our performance and management saving us money. 

What's my experience with pricing, setup cost, and licensing?

There were budgetary constraints that keep us from investing in the single pane of glass traffic management feature. We saw a demo of this feature about a year to a year and a half ago.

Which other solutions did I evaluate?

We have not looked into any other options.

What other advice do I have?

Go for it. 

It's always better to go with a nimble, growing company. Partner your growth with their journey. It's always beneficial for a stable setup. 

Disclosure: IT Central Station contacted the reviewer to collect the review and to validate authenticity. The reviewer was referred by the vendor, but the review is not subject to editing or approval by the vendor.
Add a Comment
Guest
Sign Up with Email