Flink performance testing

[mahesh salunkhe]

Team, 

What are the framework I should be using for Flink End-to-end Performance Testing? 

I would like to test performance of each flink operators, back pressure etc   

[Piotr Nowojski-4]

Hi,

I'm not sure what you are asking for. We do not provide benchmarks for all of the operators. We currently have a couple of micro benchmarks [1] for some of the operators, and we are also setting up some adhoc benchmarks when implementing various features. If you want to benchmark something besides micro benchmarks [1], you would need to set up the benchmarks on your own.

Piotrek

[1] https://github.com/apache/flink-benchmarks

śr., 16 wrz 2020 o 17:42 mahesh salunkhe <[hidden email]> napisał(a):

Team, 

What are the framework I should be using for Flink End-to-end Performance Testing? 

I would like to test performance of each flink operators, back pressure etc   

[mahesh salunkhe]

I would like to do performance testing for my flink job specially related with volume, how my flink job perform if more streaming data coming to my source connectors and measure benchmark for various operators?          

On Wed, 16 Sep 2020 at 12:03, Piotr Nowojski <[hidden email]> wrote:

Hi,

I'm not sure what you are asking for. We do not provide benchmarks for all of the operators. We currently have a couple of micro benchmarks [1] for some of the operators, and we are also setting up some adhoc benchmarks when implementing various features. If you want to benchmark something besides micro benchmarks [1], you would need to set up the benchmarks on your own.

Piotrek

[1] https://github.com/apache/flink-benchmarks

śr., 16 wrz 2020 o 17:42 mahesh salunkhe <[hidden email]> napisał(a):

Team, 

What are the framework I should be using for Flink End-to-end Performance Testing? 

I would like to test performance of each flink operators, back pressure etc   

[Piotr Nowojski-4]

Hi,

But what are you asking for?

Is it possible to do such benchmarks? Yes, it is possible. People are doing it all the time. Start a cluster, feed the data, measure the throughput (either via custom diagnostic operators, or via metrics [1]). 

Is there some framework to do it? Not that I know of. Are there tools that would help with that? Yes plenty. Basically anything that would help you to automate cluster deployment and metrics collection.

If you want to setup micro benchmarks on a single machine, you could find inspiration how can it be done in flink-benchmarks [2] and DataStream API benchmark in particular [3]

Piotrek

[1] https://ci.apache.org/projects/flink/flink-docs-stable/monitoring/metrics.html

[2] https://github.com/apache/flink-benchmarks

[3] https://github.com/apache/flink-benchmarks/tree/master/src/main/java/org/apache/flink/benchmark

śr., 16 wrz 2020 o 19:35 mahesh salunkhe <[hidden email]> napisał(a):

I would like to do performance testing for my flink job specially related with volume, how my flink job perform if more streaming data coming to my source connectors and measure benchmark for various operators?          

On Wed, 16 Sep 2020 at 12:03, Piotr Nowojski <[hidden email]> wrote:

Hi,

I'm not sure what you are asking for. We do not provide benchmarks for all of the operators. We currently have a couple of micro benchmarks [1] for some of the operators, and we are also setting up some adhoc benchmarks when implementing various features. If you want to benchmark something besides micro benchmarks [1], you would need to set up the benchmarks on your own.

Piotrek

[1] https://github.com/apache/flink-benchmarks

śr., 16 wrz 2020 o 17:42 mahesh salunkhe <[hidden email]> napisał(a):

Team, 

What are the framework I should be using for Flink End-to-end Performance Testing? 

I would like to test performance of each flink operators, back pressure etc