How to parallelize SDPJ?

[c0ldfish96]

I am familiar with SDPB and I just want to have a try on SDPJ. I want to know how to parallelize SDPJ. From the perspective of CPU usage, it seems that directly calling the 'sdp' function does not appear to parallelize. I run the code by using the Juptyer Notebook on Windows. By the way, is it easy to use run SDPJ on the cluster? It will be helpful to say something about running SDPJ on the cluster since I spent a lot of time installing SDPB on the cluster.

[FishboneChiang]

Hello c0ldfish96,

Thanks for reaching out! Glad to know that someone is actually using my program. To parallelize, simply run

export JULIA_NUM_THREADS=4

before entering Jupyter Notebook. This will automatically enable multithreading. Another way is to run your source code with

julia fileName.jl -t NUM_OF_THREADS

You might find this manual page helpful:

https://docs.julialang.org/en/v1/manual/multi-threading/

Please don't hesitate to let me know if you encounter any further issues and have fun bootstrapping!