- Medium: It contributes to significant difficulty to complete my task, but I can work around it.
After calling ray.init(), I can see amount of memory usage increases (250MB to 550MB depending how many cpus I start with) for the system. I have excluded dashboard by setting include_dashboard to False. Are there any other config or customized build I can try to minimize the memory overhead here as memory is pretty limited in our system? I’m only using Ray Core for intra-node inter-process communication. Any suggestion would be appreciated. Thanks!