Thank you very very very much for your patiently answers. I got it, this is
very helpful to understand the auto-parallelism Optimization.
At 2016-02-29 12:50:04, "Rajesh Balamohan" wrote:
"tez.shuffle-vertex-manager.desired-task-input-size" - Determines the amount of
desired task input siz
"tez.shuffle-vertex-manager.desired-task-input-size" - Determines the
amount of desired task input size per reduce task. Default is around 100 MB.
"tez.shuffle-vertex-manager.min-task-parallelism" - Min task parallelism
that ShuffleVertexManager should honor. I.e, if the client has set it as
100,
Hello, Respected experts:
Recently, I am studying tez reduce auto-parallelism, I read the article
"Apache Tez: Dynamic Graph Reconfiguration",TEZ-398 and HIVE-7158.
I found the HIVE-7158 said that "Tez can optionally sample data from a fraction
of the tasks of a vertex and use that informati
