accumulo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From keith-turner <...@git.apache.org>
Subject [GitHub] accumulo pull request: ACCUMULO-4229 BatchWriter Locator cache out...
Date Fri, 29 Apr 2016 18:11:56 GMT
Github user keith-turner commented on a diff in the pull request:

    https://github.com/apache/accumulo/pull/96#discussion_r61621905
  
    --- Diff: core/src/main/java/org/apache/accumulo/core/client/impl/SyncingTabletLocator.java
---
    @@ -0,0 +1,118 @@
    +/*
    + * 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.accumulo.core.client.impl;
    +
    +import java.util.Collection;
    +import java.util.List;
    +import java.util.Map;
    +
    +import org.apache.accumulo.core.client.AccumuloException;
    +import org.apache.accumulo.core.client.AccumuloSecurityException;
    +import org.apache.accumulo.core.client.Instance;
    +import org.apache.accumulo.core.client.TableNotFoundException;
    +import org.apache.accumulo.core.data.KeyExtent;
    +import org.apache.accumulo.core.data.Mutation;
    +import org.apache.accumulo.core.data.Range;
    +import org.apache.accumulo.core.security.Credentials;
    +import org.apache.hadoop.io.Text;
    +
    +/**
    + * Syncs itself with the static collection of TabletLocators, so that when the server
clears it, it will automatically get the most up-to-date version. Caching
    + * TabletLocators locally is safe when using SyncingTabletLocator.
    + */
    +public class SyncingTabletLocator extends TabletLocator {
    +  /*
    +   * Implementation: the invalidateCache calls do not need to call syncLocator because
it is okay if an outdated locator is invalidated. In this case, on the
    +   * next call to a meaningful function like binRanges, the new locator will replace
the current one and be used instead. The new locator is fresh; it has no
    +   * cache at all when created.
    +   */
    +
    +  /**
    +   * Used to tell SyncingTabletLocator how to fetch a new locator when the current one
is outdated. For example,
    +   * <code>TabletLocator.getLocator(instance, tableId);</code>
    +   */
    +  public interface GetLocatorFunction {
    +    TabletLocator getLocator();
    +  }
    +
    +  private volatile TabletLocator locator;
    +  private final GetLocatorFunction getLocatorFunction;
    +
    +  public SyncingTabletLocator(GetLocatorFunction getLocatorFunction) {
    +    this.getLocatorFunction = getLocatorFunction;
    +    this.locator = getLocatorFunction.getLocator();
    +  }
    +
    +  public SyncingTabletLocator(final Instance instance, final Text tableId) {
    +    this(new GetLocatorFunction() {
    +      @Override
    +      public TabletLocator getLocator() {
    +        return TabletLocator.getLocator(instance, tableId);
    +      }
    +    });
    +  }
    +
    +  private void syncLocator() {
    +    if (!locator.isValid())
    --- End diff --
    
    Two style comments, feel free to ignore.  Just things I thought of while looking at the
code.
    
    Could use an atomic reference instead of synchronized, but thats just a matter of style.
 The synchronized code does seem correct.
    
    Unrelated I would have  a getter function that does the freshness check, this way the
locator volatile only needs to be read once.  Current code reads the locator volatile in `syncLocator()`
and then immediately after `syncLocator()` it reads the volatile again.  I doubt 2 round trips
to main memory instead of three makes a performance difference in the big scheme of things,
its another stylistic comment.
    
    ```java
    private TabletLocator getLocator() {
        TabletLocator localLocator = locator.get(); // in the normal case only read locator
volatile once, if all internal code calls getLocator()
        if(!localLocator.isValid()){
           TabletLocator nl = getNewLocator();
           if(locator.compareAndSet(localLocator, nl)){
              localLocator = nl;
           } else {
              localLocator = locator.get();
           }
        }
        
       return localLocator;
    }
    ```


---
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 infrastructure@apache.org or file a JIRA ticket
with INFRA.
---

Mime
View raw message