Re: ExecutionContext, batch ordering clarification

2022-07-19 Thread Yaron Gvili
2022 10:59 AM To: dev@arrow.apache.org Subject: Re: ExecutionContext, batch ordering clarification Thanks Weston, two follow up questions: (1) What is the threading model when passing "exector=nullptr" to "ExecContext" ? (Does it only uses one thread?) (2) For the file rea

Re: ExecutionContext, batch ordering clarification

2022-07-19 Thread Li Jin
Thanks Weston, two follow up questions: (1) What is the threading model when passing "exector=nullptr" to "ExecContext" ? (Does it only uses one thread?) (2) For the file reader, if we want to ensure batches coming out of the reader are ordered but also have parallelism, I'd imagine doing sth like

Re: ExecutionContext, batch ordering clarification

2022-07-19 Thread Weston Pace
If you are using a source node (which it appears you are) then it will be creating new thread tasks for each batch. So, in theory, these could get out of order. My guess is that the file reader is slow enough that by the time you load batch N from disk and decode it, you have a pretty good chance