We performed a comparison between A10 Networks Thunder ADC, HAProxy, and Loadbalancer.org based on real PeerSpot user reviews.
Find out what your peers are saying about F5, Citrix, HAProxy and others in Application Delivery Controllers (ADC)."The ADCs are pretty straightforward and easy to use. There is a GUI base where you can go in and see everything, but they also have a CLI base where you can use a command and get the information that you want, very fast."
"The DNS application firewall and load balancing are very valuable."
"Compared to F5, which I used about six years ago, the A10 is much easier when routing. You don't have to use the wildcard bits to route it between the different segments. It's much less troublesome to configure."
"The solution is flexible."
"We have two appliances and I'm able to move my application from one appliance to another. I don't have to move my whole A10 to be active on the other side or to be passive on the other side. If an application is having a problem, I can just move it using a command."
"A lot of our SSL management is done on the front-end side, so there is one pane of glass for a lot of our security certificates. It gives us visibility. It also falls under when certificates are going to expire. Even for servers that are coming down, we can see how that affects the traffic flow by using the services map."
"The Deterministic CGNAT feature is valuable for us."
"The solution is stable."
"I can simplify configurations of many internal services (e.g. Web server configs) by moving some elements (like SSL) to HAProxy. I can also disable additional applications, like Varnish, by moving traffic shaping configurations to HAProxy."
"I have found HAProxy very helpful in replicating production environment architecture in a development and testing environment."
"Stability is number one."
"We were able to use HAProxy for round robin with our databases, or for a centralized TCP connection in one host."
"Scalable and inexpensive."
"The anti-DDOS PacketShield filtering solution (embedded in the physical appliances) as well as the BGP route injection are great features and heavily used."
"It has allowed us to evenly distribute the load across a number of servers, and check their health and automatically react to errors."
"Tech support is super-quick to respond, and always on target with answers specific to the current issue."
"The user interface precludes need to be well versed with Linux IPVS command line. This make it easy for junior team members to participate in managing load balancing needs."
"Existing customers are trying to migrate from the physical F5 load balancer to the AVI load balancer because it is scalable and easily managed."
"The connection that this solution helps our servers maintain has been most useful."
"I found scalability in Loadbalancer.org valuable."
"It helps us to route the traffic to the available servers. If we didn't have Loadbalancer we would fail to set the end-user and it would cause a failure in the cluster."
"For now, it's stable."
"The performance is good."
"We have about 30,000 connections going through at any one time and it's fine, it doesn't seem to sweat. It doesn't get overloaded."
"The user interface is not as pretty as it could be."
"We are starting to do a lot with containers and how the solution hooks into Kubernetes that we haven't explored. I'm hoping that they have a lot of hooks into Kubernetes. That would be the part for improvement: Marketing use cases with containers."
"The solution should add automation features in the next release."
"They need to make the user interface (GUI) a bit more usable and intuitive. Some features can be a little difficult to find at times. Sometimes, the workflow in the GUI doesn't match the workflow of an actual workflow. E.g., if I want to create a load balancer application, sometimes you've got to do things a bit out of order in the GUI in order to make it work right."
"The setup depends on certain situations. In certain scenarios, it may be more complex than others. For example, while the initial configuration may be easy, the environment itself may be complex and that may limit the ease of deployment. It is easy for those who understand their environment."
"The solution does logging, but the logging capacity is really small. Because we have a bunch of traffic here, we usually get a logging-side warning that "This many logs were lost because of the heavy traffic." If the logging was better, that would be very good."
"The costs can be quite high."
"There are competitors that have more features."
"The web stats UI, which provides the status of the health and numbers, could greatly benefit from having a RESTful interface to control the load-balanced nodes. Although there is a hack around the UI (by issuing a POST request to HAProxy with parameters), a RESTful interface would greatly improve the automation process (through Chef and Ansible)."
"Dynamic update API. More things should be possible to be configured during runtime."
"If nbproc = 2, you will have two processes of HAProxy running. However, the stats of HAProxy will not be aggregated, meaning you don't really know the collective status in a single point of view."
"There is room for improvement in the pricing model. It could be cheaper."
"The solution can be improved by controlling TCP behavior better and mandating to clients what the expected outcome must be in order to avoid receiving contestant timeout logs."
"We would like to see dynamic ACL and port update support. Our infrastructure relies on randomly allocated ports and this feature would allow us to update without restarting the process."
"HAProxy could improve by making the dashboards easier to use, and better reports and administration tickets."
"The visibility could be improved."
"Originally we had some stability issues with it, so they replaced it with a new box and it's fine."
"The interface from Loadbalancer.org should be improved."
"The configuration is somewhat complicated. Someone who does not know the solution may find this challenging."
"If I have to say something, I suppose they could add an automated configuration backup to an FTP location (or something similar) so you don’t have to manually do it. I don’t see this as a problem, of course, as the configuration rarely changes and we only need one backup, but maybe for other users that feature would be handy."
"It doesn't have the bonding capability feature."
"The automatic refresh of the System Overview webpage: It sometimes has an extra webpage reload (after a change) before you see it is executed. This can be confusing."
"The solution can be a bit pricey."
"There are many features you can set in the backend of Loadbalancer. They should simplify the configuration. The administrator should be able to configure it more simply. How it is now, you can only configure it if you have a lot of experience."
