HIVE-17669: Cache to optimize SearchArgument deserialization (Mithun Radhakrishnan reviewed by Prasanth Jayachandran)
Project: http://git-wip-us.apache.org/repos/asf/hive/repo Commit: http://git-wip-us.apache.org/repos/asf/hive/commit/c332103e Tree: http://git-wip-us.apache.org/repos/asf/hive/tree/c332103e Diff: http://git-wip-us.apache.org/repos/asf/hive/diff/c332103e Branch: refs/heads/branch-2 Commit: c332103ef3254f15dca1077335191ecd96e65c38 Parents: 58fa65a Author: Mithun Radhakrishnan <mit...@apache.org> Authored: Thu Oct 12 10:41:38 2017 -0700 Committer: Mithun Radhakrishnan <mit...@apache.org> Committed: Thu Oct 12 10:41:38 2017 -0700 ---------------------------------------------------------------------- .../org/apache/hadoop/hive/conf/HiveConf.java | 3 + .../hive/ql/io/sarg/ConvertAstToSearchArg.java | 76 ++++++++++++++++++-- 2 files changed, 74 insertions(+), 5 deletions(-) ---------------------------------------------------------------------- http://git-wip-us.apache.org/repos/asf/hive/blob/c332103e/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java ---------------------------------------------------------------------- diff --git a/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java b/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java index 62eabbc..3d3f3b2 100644 --- a/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java +++ b/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java @@ -1382,6 +1382,9 @@ public class HiveConf extends Configuration { "references for the cached object. Setting this to true can help avoid out of memory\n" + "issues under memory pressure (in some cases) at the cost of slight unpredictability in\n" + "overall query performance."), + HIVE_IO_SARG_CACHE_MAX_WEIGHT_MB("hive.io.sarg.cache.max.weight.mb", 10, + "The max weight allowed for the SearchArgument Cache. By default, the cache allows a max-weight of 10MB, " + + "after which entries will be evicted."), HIVE_LAZYSIMPLE_EXTENDED_BOOLEAN_LITERAL("hive.lazysimple.extended_boolean_literal", false, "LazySimpleSerde uses this property to determine if it treats 'T', 't', 'F', 'f',\n" + http://git-wip-us.apache.org/repos/asf/hive/blob/c332103e/ql/src/java/org/apache/hadoop/hive/ql/io/sarg/ConvertAstToSearchArg.java ---------------------------------------------------------------------- diff --git a/ql/src/java/org/apache/hadoop/hive/ql/io/sarg/ConvertAstToSearchArg.java b/ql/src/java/org/apache/hadoop/hive/ql/io/sarg/ConvertAstToSearchArg.java index 997334b..2de8319 100644 --- a/ql/src/java/org/apache/hadoop/hive/ql/io/sarg/ConvertAstToSearchArg.java +++ b/ql/src/java/org/apache/hadoop/hive/ql/io/sarg/ConvertAstToSearchArg.java @@ -21,13 +21,18 @@ package org.apache.hadoop.hive.ql.io.sarg; import java.sql.Date; import java.sql.Timestamp; import java.util.List; +import java.util.concurrent.Callable; +import java.util.concurrent.ExecutionException; +import com.google.common.cache.Cache; +import com.google.common.cache.CacheBuilder; +import com.google.common.cache.Weigher; import org.apache.commons.codec.binary.Base64; import org.apache.hadoop.conf.Configuration; import org.apache.hadoop.hive.common.type.HiveChar; import org.apache.hadoop.hive.common.type.HiveDecimal; +import org.apache.hadoop.hive.conf.HiveConf; import org.apache.hadoop.hive.ql.exec.SerializationUtilities; -import org.apache.hadoop.hive.ql.io.sarg.LiteralDelegate; import org.apache.hadoop.hive.ql.plan.ExprNodeColumnDesc; import org.apache.hadoop.hive.ql.plan.ExprNodeConstantDesc; import org.apache.hadoop.hive.ql.plan.ExprNodeDesc; @@ -436,14 +441,75 @@ public class ConvertAstToSearchArg { } } - public static final String SARG_PUSHDOWN = "sarg.pushdown"; + private static volatile Cache<String, SearchArgument> sargsCache = null; + + private static synchronized Cache<String, SearchArgument> initializeAndGetSargsCache(Configuration conf) { + if (sargsCache == null) { + sargsCache = CacheBuilder.newBuilder() + .weigher(new Weigher<String, SearchArgument>() { + @Override + public int weigh(String key, SearchArgument value) { + return key.length(); + } + }) + .maximumWeight( + HiveConf.getIntVar(conf, + HiveConf.ConfVars.HIVE_IO_SARG_CACHE_MAX_WEIGHT_MB) * 1024 *1024 + ) + .build(); // Can't use CacheLoader because SearchArguments may be built either from Kryo strings, + // or from expressions. + } + return sargsCache; + } + + private static Cache<String, SearchArgument> getSargsCache(Configuration conf) { + return sargsCache == null? initializeAndGetSargsCache(conf) : sargsCache; + } + + private static boolean isSargsCacheEnabled(Configuration conf) { + return HiveConf.getIntVar(conf, HiveConf.ConfVars.HIVE_IO_SARG_CACHE_MAX_WEIGHT_MB) > 0; + } + + private static SearchArgument getSearchArgumentFromString(Configuration conf, final String sargString) { + + try { + return isSargsCacheEnabled(conf)? + getSargsCache(conf).get(sargString, new Callable<SearchArgument>() { + @Override + public SearchArgument call() { + return create(sargString); + } + }) + : create(sargString); + } + catch (ExecutionException exception) { + throw new RuntimeException(exception); + } + } + + private static SearchArgument getSearchArgumentFromExpression(final Configuration conf, final String sargString) { + + try { + return isSargsCacheEnabled(conf)? + getSargsCache(conf).get(sargString, new Callable<SearchArgument>() { + @Override + public SearchArgument call() { + return create(conf, SerializationUtilities.deserializeExpression(sargString)); + } + }) + : create(conf, SerializationUtilities.deserializeExpression(sargString)); + } + catch (ExecutionException exception) { + throw new RuntimeException(exception); + } + } + public static SearchArgument create(Configuration conf, ExprNodeGenericFuncDesc expression) { return new ConvertAstToSearchArg(conf, expression).buildSearchArgument(); } - private final static ThreadLocal<Kryo> kryo = new ThreadLocal<Kryo>() { protected Kryo initialValue() { return new Kryo(); } }; @@ -459,9 +525,9 @@ public class ConvertAstToSearchArg { public static SearchArgument createFromConf(Configuration conf) { String sargString; if ((sargString = conf.get(TableScanDesc.FILTER_EXPR_CONF_STR)) != null) { - return create(conf, SerializationUtilities.deserializeExpression(sargString)); + return getSearchArgumentFromExpression(conf, sargString); } else if ((sargString = conf.get(SARG_PUSHDOWN)) != null) { - return create(sargString); + return getSearchArgumentFromString(conf, sargString); } return null; }