Tao He
1 week agoHi folks, I have read [the excellent tutorial about distributed training](https://docs.flyte.org/projects/cookbook/en/stable/auto/case_studies/ml_training/spark_horovod/keras_spark_rossmann_estimator.html#). I'm wondering whether the sentence "'SparkBackend' uses
to execute the distributed training function" means that flyte will launchhorovod.spark.run
spark workers onnum_proc
flyte worker? Or it just launches them on one spark worker?num_proc
Ketan (kumare3)
1 week agohey @Tao He welcome to the community. There are no
today. Flyte launches ephemeral Spark clusters using Spark operator or Spark for K8s. In the horovod case, it simply uses the horovod with spark integration, just make sure the configuration is correctFlyte workersTao He
1 week agoFlyte launches ephemeral Spark clusters using Spark operator or Spark for K8s. I see. Thanks!
- Hi @Ketan (kumare3), I have a follow question. Under such a setting the spark worker still fetches data (the flyte literals) using the flyte API from the underlying remote object store(s3) ? In this case I'm wondering how the flyte's local cache works, as the spark worker is not managed by flyte and the dataframe that passed to the spark shouldn't present in the local cache. Or, should I understand the case as the local cache is to avoid duplicating fetching data (the dataframe that passed to the spark) from s3, but the first fetch is still needed?
Ketan (kumare3)
1 week agoI do not follow- what is local cache? Do you mean Flyte cache for tasks. That is only used to memorize a previous execution and prevent a re run. There is no optimization for spark handling the data. Spark will have to download from object store Flyte tries to prevent starting the spark cluster itself, if the task is cachedTao He
1 week agoFlyte tries to prevent starting the spark cluster itself, if the task is cached Now I can understand that the spark job (launching the spark cluster using spark-operator and submit a job to it) is "a single task" in a flyte workflow.
- I have follow-up question. Will flyte's scheduler consider the co-locality between two tasks inside a workflow? After some experiments now I can see the answer should be "won't". As the flyte propeller launches pods for each execution. Corrected me if I was wrong.
Ketan (kumare3)
1 week agoYa not today, but this is feature plannedTao He
1 week agoHappy to know that. May I find any related docs or enhancement proposal draft from somewhere on this topic? Thanks!Ketan (kumare3)
1 week agoYes will share the enhancement proposalTao He
1 week agoLooking forward to it!Ketan (kumare3)
1 week ago@Tao He phase 2 of this rfc, cc @Kevin Su can we please split this into 2 rfcs and add to the official rfcs? https://docs.google.com/document/d/1-695lxz8a-GFz4cFamGkF1NhspKMkDGIDdkY9EDLMz8/editTao He
1 week agoThanks for sharing ❤️Ketan (kumare3)
1 week agoAbsolutely- Ya ray is almost integrated and we will eventually work on cluster reuse
Join thread in Slack
View count: 2