ulysses-you commented on code in PR #5439: URL: https://github.com/apache/incubator-gluten/pull/5439#discussion_r1590476656
########## gluten-core/src/main/java/org/apache/gluten/memory/memtarget/spark/DynamicOffHeapSizingPolicyChecker.java: ########## @@ -0,0 +1,71 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ +package org.apache.gluten.memory.memtarget.spark; + +import org.apache.gluten.GlutenConfig; + +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +import java.util.concurrent.atomic.AtomicLong; + +public final class DynamicOffHeapSizingPolicyChecker { + private static final Logger LOG = + LoggerFactory.getLogger(DynamicOffHeapSizingPolicyChecker.class); + private final long maxOnHeapMemoryInBytes = GlutenConfig.getConf().onHeapMemorySize(); + private final AtomicLong usedOffHeapBytes = new AtomicLong(); + + DynamicOffHeapSizingPolicyChecker() {} + + public boolean canBorrow(long size) { Review Comment: Oversubscribe memory is quite risk. I think we should at least respect the Spark idea. The max execution memory in Spark is: ``` (spark.executor.memory - 300MB) * spark.memory.fraction ``` `totalMemory() - freeMemory()` breaks this idea. Even if the execution offload to native, there are still many things depend on on-heap memory, e.g., the metadata(plan, split, map status, event...), broadcast block, rdd cache. One thought is that, we'd better to follow the Spark, and change it to: ``` (totalMemory() - freeMemory() - 300MB) * spark.memory.fraction ``` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
