HBase compaction分析

如下代碼都是基於HBase-0.98版本。

compaction是將多個HFile合併爲一個HFile操做，

進行compaction有以下幾個做用：

減小HFile文件的個數，

HFile減小可能提升讀取性能，

清除過時和刪除數據。

compaction有二種類型：minor 和 major

1）Minor操做只用來作部分文件的合併操做以及包括minVersion=0而且設置ttl的過時版本清理，不作任何刪除數據、多版本數據的清理工做。

2）Major操做是對Region下的HStore下的全部StoreFile執行合併操做，會作刪除數據，多版本數據清理工做，最終的結果是整理合並出一個文件。

HRegionServer在啓動的時候，會啓動compactionChecker線程，compactionChecker會檢測region是否須要compaction。

主要執行的邏輯以下：

protected void chore() {
      for (HRegion r : this.instance.onlineRegions.values()) {
        if (r == null)
          continue;
        for (Store s : r.getStores().values()) {
          try {
            long multiplier = s.getCompactionCheckMultiplier();
            assert multiplier > 0;
            if (iteration % multiplier != 0) continue;
            if (s.needsCompaction()) {
              // Queue a compaction. Will recognize if major is needed.
              this.instance.compactSplitThread.requestSystemCompaction(r, s, getName()
                  + " requests compaction");
            } else if (s.isMajorCompaction()) {
              if (majorCompactPriority == DEFAULT_PRIORITY
                  || majorCompactPriority > r.getCompactPriority()) {
                this.instance.compactSplitThread.requestCompaction(r, s, getName()
                    + " requests major compaction; use default priority", null);
              } else {
                this.instance.compactSplitThread.requestCompaction(r, s, getName()
                    + " requests major compaction; use configured priority",
                  this.majorCompactPriority, null);
              }
            }
          } catch (IOException e) {
            LOG.warn("Failed major compaction check on " + r, e);
          }
        }
      }
      iteration = (iteration == Long.MAX_VALUE) ? 0 : (iteration + 1);
    }
  }

如代碼，遍歷onlineRegions，獲取每一個region的Store進行判斷，其中needsCompaction邏輯以下：

public boolean needsCompaction(final Collection<StoreFile> storeFiles,
      final List<StoreFile> filesCompacting) {
    int numCandidates = storeFiles.size() - filesCompacting.size();
    return numCandidates >= comConf.getMinFilesToCompact();
  }

minFilesToCompact由hbase.hstore.compaction.min(老版本是：hbase.hstore.compactionThreshold)控制，默認值爲3，即store下面的storeFiles數量 減去 正在compaction的數量 >=3是，須要作compaction。

needsCompaction爲true後，會調用compactSplitThread.requestSystemCompaction方法，發送compaction請求，具體內容會在compactSplitThread線程裏分析。

needsCompaction爲false後，會判斷是否爲isMajorCompaction，具體邏輯以下：

/*
   * @param filesToCompact Files to compact. Can be null.
   * @return True if we should run a major compaction.
   */
  public boolean isMajorCompaction(final Collection<StoreFile> filesToCompact)
      throws IOException {
    boolean result = false;
    long mcTime = getNextMajorCompactTime(filesToCompact);
    if (filesToCompact == null || filesToCompact.isEmpty() || mcTime == 0) {
      return result;
    }
    // TODO: Use better method for determining stamp of last major (HBASE-2990)
    long lowTimestamp = StoreUtils.getLowestTimestamp(filesToCompact);
    long now = System.currentTimeMillis();
    if (lowTimestamp > 0l && lowTimestamp < (now - mcTime)) {
      // Major compaction time has elapsed.
      long cfTtl = this.storeConfigInfo.getStoreFileTtl();
      if (filesToCompact.size() == 1) {
        // Single file
        StoreFile sf = filesToCompact.iterator().next();
        Long minTimestamp = sf.getMinimumTimestamp();
        long oldest = (minTimestamp == null)
            ? Long.MIN_VALUE
            : now - minTimestamp.longValue();
        if (sf.isMajorCompaction() &&
            (cfTtl == HConstants.FOREVER || oldest < cfTtl)) {
          if (LOG.isDebugEnabled()) {
            LOG.debug("Skipping major compaction of " + this +
                " because one (major) compacted file only and oldestTime " +
                oldest + "ms is < ttl=" + cfTtl);
          }
        } else if (cfTtl != HConstants.FOREVER && oldest > cfTtl) {
          LOG.debug("Major compaction triggered on store " + this +
            ", because keyvalues outdated; time since last major compaction " +
            (now - lowTimestamp) + "ms");
          result = true;
        }
      } else {
        if (LOG.isDebugEnabled()) {
          LOG.debug("Major compaction triggered on store " + this +
              "; time since last major compaction " + (now - lowTimestamp) + "ms");
        }
        result = true;
      }
    }
    return result;
  }

未完待續。。。