Most Helpful Review
We asked business professionals to review the solutions they use. Here are some excerpts of what they said:
The ease of use is very good. It's very robust. It just sits and works.
A10 explained why the latency dropped significantly on a site that we have.
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.
We do have the option of creating virtual chassis, so that gives it a bit more security. If we find an application which is not going to play well in the main pool, we can easily create a virtual chassis and have that application in that virtual chassis. With the virtual chassis we can also create system partitions and have a test system for test applications, and have the others elsewhere.
The Global Server Load Balancing (GSLB) is simple to use.
With the Thunder SSLi, we're better protected. We can stop use of VPN and proxies. We are better protected against dirty traffic coming back to our schools. Having a secure decrypt zone with the equipment lowers the chances that our security infrastructure could possibly miss an attack.
Being a public entity and having a public website which is highly visible with a lot of traffic, we are a target for DDoS. Within the last year, we have had a couple of DDoS attacks which could have affected our web traffic and taken down certain parts of our website. This did not happen because the A10 was able to mitigate the attacks using rate limiting that can be configured for DDoS mitigation on the box.
It is very useful to have a simple dashboard where you can login and look into what your traffic patterns are, then look and see what times of day you're experiencing the heaviest traffic. You can quickly identify if you are possibly having a security issue or security breach. It makes it very easy to use the box.
It solves a problem for me where I can build files, not based on the health of the check, but rather the speed of the check.
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.
Advanced traffic rules, including stick tables and ACLs, which allow me to shape traffic while it's load balanced.
Performance configuration options with threads, processes, and core stickiness are very valuable.
The ability to handle a sequence of front- and back-ends gives the user the opportunity to send traffic through different services.
We were able to use HAProxy for round robin with our databases, or for a centralized TCP connection in one host.
It reduced the load on our main load balancers.
We did not need technical support because the documentation is good.
The user interface is what people complain about most of the time, particularly if they don't use it very often. Then they complain that it's a bit clunky.
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.
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.
Traffic flow issues are very difficult, as there's no means for us to analyze the traffic coming in or out of the appliance without technical support.
I would like them to have a better UI (better universal design).
When it comes to support, there is always room for improvement. First call resolution is not always there for urgent issues. The first call resolution is something that could be improved upon.
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 logging functionality could use improvement, as it is a little cryptic.
There are three main areas to improve: 1) Make remote management more modern by adding API. 2) Propose a general HA solution for HAProxy (no I'm using keepalived for this). 3) Thread option should be a bit more stable.
The configuration should be more friendly, perhaps with a Web interface. For example, I work with the ClusterControl product for Severalnines, and we have a Web interface to deploy the HAProxy load-balancer.
We need to handle new connections by dropping, or queuing them while the HAProxy restarts, and because HAProxy does not handle split config files.
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).
I would like to evaluate load-balancing algorithms other than round robin and SSL offloading. Also, it would be helpful if I could logically divide the HAProxy load-balancing into multiple entities so that I would install one HA Proxy LB application which could be used for different Web servers for different applications. I am not sure if these features are available.
They should introduce one feature that I know many people, including me, are waiting for: HAProxy should have provide hot-swipe for back-end servers. Also, they need a more detailed GUI for monitoring and configuration.
It needs proper HTTP/2 support.
Pricing and Cost Advice
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.
It is $7000 per unit for the support annually.
We previously had F5 and switched because of costs.
We always pay for support. Any organization of our size who doesn't is asking for problems.
When you purchase the equipment, you purchase the licensing and warranty. It's all fairly standard. We haven't been caught with anything surprising.
One of the main reasons for switching away from Cisco was the licensing model. A10 gives you global server load balancing for free, while Cisco charged a significant licensing fee for that.
For the hardware and license, we paid $35,000 per box, which was a one-time cost. Then, for the Gold Support on the two boxes, we pay $9400 annually.
The pricing is a third of the price of the F5 competition.
The only cost is for the image manager, who is responsible for uploading the image, and that is trivial.
I think that the pricing is very fair, I would definitely recommend buying the Enterprise license.
We use NGINX as well. However, because the health checks are a paid feature, I like to avoid it whenever possible.
If you don't have expertise then go with the licensed version. Otherwise, open-source is the best solution.
Very good value for the money. One of the simplest licensing schemes in this category of products.
Test/lab virtual machines can be installed without a licence. They can't be used for performance testing but otherwise behave like production nodes.
The price is well worth it. HAProxy Enterprise Edition paid for itself within months, simply due to the resiliency it brings. It was a bit more expensive than we were originally interested in paying, but we are thankful we chose to go with HAProxy.
HAProxy is free software. There are optional paid products (support/appliances).
out of 27 in Application Delivery Controllers
Average Words per Review
out of 27 in Application Delivery Controllers
Average Words per Review
Compared 55% of the time.
Compared 11% of the time.
Compared 6% of the time.
Compared 41% of the time.
Compared 25% of the time.
Compared 9% of the time.
Also Known As
|Thunder ADC, AX Series||HAProxy Community Edition, HAProxy Enterprise Edition, HAPEE|
A10 Networks' application networking, load balancing and DDoS protection solutions accelerate and secure data center applications and networks of thousands of the world's largest enterprises, service providers, and hyper scale web providers.
HAProxy is the most widely used software load balancer and application delivery controller in the world. The core HAProxy application delivery engine is an open source project chiefly maintained by HAProxy Technologies and assisted by a thriving open source community. HAProxy Community Edition is available for free at haproxy.org. HAProxy Enterprise Edition is packaged with additional enterprise class features, services and premium support from HAProxy Technologies.
Learn more about A10 Networks Thunder ADC
Learn more about HAProxy
|123inkt.nl, Bentley University, Box, Brainshark, Buienradar, Capgemini, CGN/LSN & NAT64, Chengdu Telecom, Club One, Code Ready, CRC Health Group, Cyso, Deutsche Telekom, Earth Class Mail, Excite, FFF Enterprises, Florence County, Framingham State University, From30||Booking.com, GitHub, Reddit, StackOverflow, Tumblr, Vimeo, Yelp|
Recreational Facilities And Services Company11%
Mining And Metals Company11%
Software R&D Company48%
Comms Service Provider10%
K 12 Educational Company Or School6%
Financial Services Firm33%
Marketing Services Firm22%
Software R&D Company24%
Comms Service Provider9%