Github user mridulm commented on a diff in the pull request: https://github.com/apache/spark/pull/16603#discussion_r96581689 --- Diff: core/src/main/java/org/apache/spark/memory/TaskMemoryManager.java --- @@ -144,8 +164,24 @@ public long acquireExecutionMemory(long required, MemoryConsumer consumer) { // spilling, avoid to have too many spilled files. if (got < required) { // Call spill() on other consumers to release memory + // Sort the consumers according their memory usage. So we avoid spilling the same consumer + // which is just spilled in last few times and re-spilling on it will produce many small + // spill files. + List<MemoryConsumer> sortedList = new ArrayList<>(consumers.size()); for (MemoryConsumer c: consumers) { if (c != consumer && c.getUsed() > 0 && c.getMode() == mode) { + sortedList.add(c); + } + } + Collections.sort(sortedList, new ConsumerComparator()); + for (int listIndex = 0; listIndex < sortedList.size(); listIndex++) { + MemoryConsumer c = sortedList.get(listIndex); + // Try to only spill on the consumer which has the required size of memory. + // As the consumers are sorted in descending order, if the next consumer doesn't have + // the required memory, then we need to spill the current consumer at least. + boolean doSpill = (listIndex + 1) == sortedList.size() || + sortedList.get(listIndex + 1).getUsed() < (required - got); + if (doSpill) { --- End diff -- I like the fact that this implementation does not need to incur the cost of remove in a TreeMap. Unfortunately, I dont think it is sufficient : the impl assumes that spill() will actually always give you back getUsed - from the rest of the code in the method, this does not look like a valid assumption to make. This can resulting in spilling a large number of smaller blocks, and potentially itself. For example: required = 500MB, consumers = 1.5GB 1GB 500MB 2MB 1MB .. If spilling 500MB resulted in (say) releasing 490MB, we might end up spilling a large number of blocks and also (potentially) end up spilling itself - also can end up returning less than requested while enough memory does exist to satisfy the request.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. --- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org