Re: push-based external shuffle service on K8S - Spark 4.0? Earlier versions?

2024-06-06 Thread Ye Zhou
Hi Ofir. Right now, the push based shuffle within Spark is only supported for Spark on YARN, with external shuffle service running as auxiliary service in NodeManager, but not natively on K8s. As far as I know, there are no recent plans to add the support for Spark on K8s natively. For question 2,

Re: push-based external shuffle service on K8S - Spark 4.0? Earlier versions?

2024-06-06 Thread Keyong Zhou
Hi Ofir, I can provide some information about use cases for Apache Celeborn. Apache Celeborn can be deployed on K8s and standalone, both are widely used in production environment by users. The largest cluster I know contains more than 1,000 Celeborn workers. Celeborn is specially beneficial for

push-based external shuffle service on K8S - Spark 4.0? Earlier versions?

2024-06-06 Thread Ofir Manor
Hi, Regarding the external shuffle service on K8S and especially the push-based variant that was merged in 3.2: 1. Are there plans to make it supported and work out-of-the-box in 4.0? 2. Did anyone make it work for themselves in 3.5 or earlier? If so, can you share your experience and what w