BlazeMeter Reviews

• The ability to quickly scale out a load test across the world, just to simulate realistic world traffic conditions.

For us in particular, we're a product that gets heavily used during the election season. We've always had trouble handling this burst in the past, so this past year, I've started using BlazeMeter at this company. We took from January to July using BlazeMeter to really model out what the traffic looked like in previous years and to be able to handle the load when November came and it went through just fine without a glitch.

At the moment, it's heavily based on putting a load on systems through API endpoints/HTTP endpoints. There is a lot of existing tooling out there in organizations that deal with a Selenium-based test and they only have a Selenium web driver option right now, which is not the nicest thing to use. Organizations currently have Selenium tests written and, like us for example, a huge suite in Java. It would be nice to be able to just throw them in there.

I’ve been using BlazeMeter for about four years.

This is a cloud-based solution. I've never really had any trouble with it.

We have not had scalability issues for our needs. We're talking about scaling stuff out to hundreds of thousands of requests per second and those in turn generate a lot more traffic internally. So with whatever we use, it's perfectly fine.

We did not use technical support.

We didn't have a previous solution. I personally came to this company in the past year and a half. I've always used BlazeMeter at my other companies and there was a big, gaping hole here. They were doing everything manually with running JMeter tests and this was just a natural fit.

For this audience, if you know JMeter, everything is pretty straightforward. You just sign up and upload your test. As far as support goes, if you do run into stuff, they have a little button on the side that lets you chat with them.

This is all in-house.

That's tough to calculate, but I can tell you that BlazeMeter helped increase quality-of-service (QoS) which in turn leads to customers having a positive experience with the product. That, of course, leads to less churn which is a good thing.

The pricing works. They're flexible. They work with you.

There was LoadStorm, loader.io, and blitz.io.

Definitely ask your engineering staff if they're familiar with JMeter. If they're not, it would be good to get a quick two-hour tutorial on it and then BlazeMeter will make a lot more sense.

We were doing load testing. We were testing web servers, and we needed to confirm that our servers were performing as well as the old ones.

One example of what happened is that we noticed that there was a significant difference between our two sites. It turned out that there was a configuration on the network side so after we fixed that, the performance was much better.

It helped to identify where there were problems, and then we could go investigate and figure out what was going on. We didn't just do one test, but this was a one-time sanity check on the new environment. One time lasted three months, but during that time, I would say we did it once a week.

There was one feature that I wanted that they didn't have. The feature was starting two tests simultaneously, but keeping the data separate, and showing separate graphs from two different test instances on the same display.

What happened is that I was trying to run two tests at the same time so I would have to do one in one window, one in the other window, and there was a few seconds difference, which was okay. Then, I would have to take the tests in one browser, and compare them to the tests in another browser.

If they could combine functionality and put them side by side with the same scale for me, that would've been ideal. I was using one standard testing configuration.

We used it, I would say, for three months, to do performance testing on a very specific application.

We did not encounter any stability issues.

We didn't have any scalability issues. I can't remember how many users we had. We weren't doing incredible load testing. We were just using heavy load testing. There were no issues on the BlazeMeter side.

I would rate the technical support as 9/10, as I used it quite a few times. There was one instance where there was a timing mixup with a support engineer, which I was disappointed about, but the service was totally great. I would call them up with a question, they would answer it. It was fine.

I did not use any solution previously. I chose them because somebody else in my company had used them before, and had told me that they were very good.

The setup was straightforward. We had a very simple setup. We had a very simple testing scenario. After I watched the one-hour video, how the thing works, and then it showed up your URLs to test, I entered that, and there were two options we had to choose. Technically, it was very straightforward just for pure load testing.

The pricing was straightforward. That was no issue. To have the features, the special features that I wanted, realistically it could've been cheaper, but you get what you pay for. It was fair pricing, it was just a lot for our purchasing pipeline.

I did a small evaluation, not very intensive. I would say a very small evaluation. Nothing significant.

I would say use it, but people also have to know that they have to do their own due diligence and they have to make sure that their servers are okay. My advice would be to gather statistics on the servers as the tests are running, so that if the test indicates that there's something wrong, you can quickly look to the CPU, to the memory, to get an idea if it looks like it's an internal issue, or if it looks like it's an external issue.

The ability to push large numbers of simulated users at our applications and having access to detailed reports: This helps us to have peace of mind that our apps can stand up to production-level loads and beyond.

We no longer have to guess whether deploying new code will affect performance or not. This ensures we are ready to meet customer demand, even at peak traffic times.

I would like to have better insights when tests fail to run properly or when the load engines occasionally break. There have been times when a test fails to start, sometimes due to a bug in my script and other times due to a problem on BlazeMeter’s side, that have required contacting support. It would be better if the interface proved enough feedback via logs, etc., for me to determine the cause and possibly fix it without needing to wait for BlazeMeter support.

I have used BlazeMeter for three years.

Occasionally, we have long wait times or trouble starting tests.

One load engine was not sufficient to run the JMeter scripts that I’ve written. Scaling up to seven load engines largely eliminated these issues.

Support is adequate. Getting initial responses for support issues can sometimes be slow.

It was very straightforward to upload the JMeter scripts and supporting files, and start running tests.

It is somewhat expensive, but the amount of testing and peace of mind it gives us is well worth the cost.

We considered creating our own internal JMeter farm, but in the end decided BlazeMeter would do a better job of end-to-end testing.

If you are looking to improve your website performance and ensure maximum uptime by load testing your system from end to end, then there is no better solution that I have found. It is extremely simple to setup.

The metrics are a great feature for analyzing each test. Being able to see the errors compiled, along with their responses, has been very useful. 

We also love having the ability to compare multiple test results in one screen. The multi-test feature is very useful as well. The ease for creating tests in BlazeMeter has been wonderful.

It has allowed us to run performance tests in order to help determine if we can scale well.

The ability to run node.js scripts needs to improve.

I have used this solution for around two months.

We have not encountered any particular stability issues.

We have not encountered any scalability issues.

We did not previously use a different solution.

The setup was pretty straightforward.

It would be nice to have more flexible pricing options.

Having a simple setup and being able to get the tests up and running right away, is the key for this solution.

Other than URL testing, there are no pieces of this software that are actually useful to us.

We were planning to load test this product for a significant release. It was somewhat helpful.

There are only certain features that actually work. Their URL testing and JMeter tests work only if testing URLs. If testing anything else, like WebDriver tests using JMeter and BlazeMeter, they do not work.

We have used it for close to two years but we are planning to NOT extend our licensing.

We had stability problems several times and we lost all our credits.

We did have scalability problems.

Technical support was OK for some situations.

We did not use a previous solution.

Setup was straightforward and simple.

I feel it’s overpriced unless they fix several things on their end.

We did evaluate other options, but we chose this one because setup was easy.

Look for your needs. This product is not a complete load-testing solution. It does offer certain types of load testing like URL testing but, other than that, we did not find them useful.

I really like the ability to spin up as many load generators as I need, and get statistics on response times and the load generators themselves. BlazeMeter support and flexibility is also hugely valuable.

To accurately test load on a webserver it takes traffic (load) from multiple “load generators”.
A load generator is a machine running load testing software (like Jmeter).
Clearly, one couldn’t accurately test 10 load balanced webservers with traffic from a standard desktop.
If one did and saw performance issues, the most likely source would be the desktop doing the testing, or the connection.
If no performance issues were seen it could be that the desktop was not able to emulate the specified load.

One blind spot of Jmeter is it does not tell the user how many threads their machine can handle. Users need to know how to monitor. Most users don’t, and set the threads higher than the machine can handle.
Blazemeter:

- Handles all that complication and makes using Jmeter simple

- Has expert staff that can point you in the right direction if your tests are not working

- Allows users to select how many load generators (machines running Jmeter) will be used during a given test.

o There are a few different contract options (1 server, 5 servers)

o Custom contracts for any number of servers you need. I really appreciate their flexibility on the contracts.

Building an in-house group of load generation JMeter servers to load test usually takes time and effort. BlazeMeter makes all that simple.

I have used JMeter for 4 years and BlazeMeter on and off for 2 to 3 years.

There were some minor issues with stability.

We did not encounter any issues here. Scalability is one of the wonderful parts of working with BlazeMeter.

Technical support has been great.

I’ve used JMeter from several load generator servers I set up to work together.

Initial setup was easy.

If you don’t see the option you need, just ask them. You’ll be able to get a custom contract. It is not free, but it is a fantastic deal compared to anything else out there.

Most recently I have been using a custom contract with Blazemeter rather than a standard month to month contract. That allows me to pay by test hours times servers used. So I can spin up as many load generators as I need. The statistics on response times and the load generators themselves are standard Blazemeter features available to anyone with a contract. BlazeMeter support has several times helped me figure out what was going on with my tests, that saved me a chunk of time. It's always nice to have some insight from a true Jmeter expert.

We considered setting up our own set of load generation JMeter servers. We also looked at HPE LoadRunner and SoapUI.

BlazeMeter is an easy way to get really accurate results from JMeters that you already have without having to build a set of networked JMeters (e.g., one with the UI running a bunch GUI-less slave machines).

Ability to generate heavy load (up to 5000 concurrent users)

Provide performance transactions summary & statistics

These features are valuable because, in my company, there is a requirement for the product to support up to 5000 concurrent users. As a performance engineer, it’s important to have a tool (like BlazeMeter) to generate heavy load and at the same time provide a report that shows how the system behaved under the load and provide performance statistics.

By using this tool, a performance engineer was able to identify performance bottlenecks, and after fixes was applied, overall product quality improved.

It would be very useful if there was a mechanism to schedule performance runs.

For example, a mechanism to automatically schedule performance tests execution, by using the calendar scheduler. This way, it would be possible to schedule multiple runs overnight, without the need to manually start each performance test.

I have used BlazeMeter with JMeter 3.0 and Java 1.8 for six months.

Stability was great; no issues.

Scalability was great; no issues.

I did not contact support.

We did not previously use a different solution.

I think prices are reasonable.

Before choosing this product, we did not evaluate other options.

It is an excellent and reliable tool. I liked it very much.

Load generation over cloud is the most valuable feature. This feature is most valuable for us because we use BlazeMeter for load testing. This helps us procure the required load generation infrastructure.

I have used BlazeMeter personally and my experience has been:

Whenever we want to provision more load generators (10 or more), it takes too long to initiate the test. Sometimes, the load generators have not even started and after waiting for 10 minutes or more, we get to know that we need to restart. This creates an awkward situation, when the client is also monitoring live execution and waiting for the test to start.

I have come across multiple instances, where the actual test was run for x hours but the server hours that were deducted from my account, were on the basis of x+1 hours. I have to keep monitoring the usage report frequently and request for credit back of all such server hours. Although, I get them back without any argument with the BlazeMeter team but still, it creates an overhead for me.

It needs to improve the server hour calculation and the provisioning of load generation machines.

I have used this solution for more than six months.

We had some stability issues; sometimes the load generators do not start at all.

We encountered some scalability issues. The start time for load test increases with the increase in number of users to be simulated.

I would give the technical support a 5/10.

The setup was straightforward.

The pricing could be more competitive.

We did not evaluate any other options.