ghz VS eCAL

I once built a quick and dirty load testing tool for a public facing service we built. The tool was pretty simple - something like https://github.com/bojand/ghz but with traffic and data patterns closer to what we expected to see in the real world. We used argo-workflows to generate scale.

One thing which we noticed was that there was a considerable difference in performance characteristics based on how we parallelized the load testing tool (multiple threads, multiple processes, multiple kubernetes pods, pods forced to be distributed across nodes).

I think that when you run non-distrubuted load tests you benefit from bunch of cool things which happen with http2 and Linux (multiplexing, resource sharing etc) which might make applications seem much faster than they would be in the real world.

I'm not sure. Maybe you can write to the discussion section of the repo https://github.com/bojand/ghz/discussions

In part 1 we looked at ghz for load testing gRPC services, and now I want to cover k6, which claims to be a modern load testing tool built for developer happiness. After only a brief experience with it I can see why is that and why Grafana moved to acquire k6 earlier this year.

It should be unbound, in this particular benchmark we set ghz concurrency to 50 and connections to 5 and we don't set the rps flag of ghz (e.g. --rps=2000, from this tool)

a second container running ghz makes unary requests to the server

New version (5.94) of the high performance interprocess / interhost communication framework eCAL is available. eCAL supports different transport layer like shared memory for interprocess and udp multicast for interhost communication. It does not force the user to use a specific serialization format but supports some of the standards like google protobuf, capnproto or flatbuffers. The API is wrapped to C, C++, Python, C#, Rust and Go (the last two I never tested ;-)). Here you can find all the documentation. Check out the great applications for monitoring, record and replay. eCAL is Apache 2 licensed, repository hosted here. Have fun :-)

I you just need a fast publish / subscribe framework for windows you can give eCAL a try. The setup is done in less a minute and their is no dependency hell. Just to mention it as alternative. Documentation here

I would concider using eCAL since it blasts everything out of the water in terms of performance and comes with some handy tooling for inspection of messages in-travel.

Source on GitHub: https://github.com/continental/ecal

Check out eCal.

This really makes me want to try https://github.com/continental/ecal with https://github.com/google/flatbuffers to see how they compare. I also know that gpc for cpp at least stops functioning by about 4 MB of request size. Which I find stupid. What if I want to send uncompressed bitmaps!

Congratulations. Well done. eCAL will update as soon as possible to the new release. The new custom header and the n:m pub/sub support are really welcome new features.

Here is their counter callback example: https://github.com/continental/ecal/blob/master/samples/cpp/counter/counter_rec_cb/src/counter_rec_cb.cpp