HDDS-13014. Improve PrometheusMetricsSink#normalizeName performance

hadoop-hdds/framework/src/main/java/org/apache/hadoop/hdds/utils/PrometheusMetricsSinkUtil.java

@@ -19,24 +19,42 @@
 
 import static org.apache.hadoop.hdds.utils.RocksDBStoreMetrics.ROCKSDB_CONTEXT_PREFIX;
 
+import com.google.common.cache.CacheBuilder;
+import com.google.common.cache.CacheLoader;
+import com.google.common.util.concurrent.UncheckedExecutionException;
 import java.util.ArrayList;
 import java.util.Collection;
 import java.util.List;
+import java.util.concurrent.ExecutionException;
 import java.util.regex.Pattern;
 import java.util.stream.Stream;
 import org.apache.commons.lang3.StringUtils;
 import org.apache.hadoop.metrics2.MetricsTag;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
 
 /**
  * Util class for
  * {@link org.apache.hadoop.hdds.server.http.PrometheusMetricsSink}.
  */
 public final class PrometheusMetricsSinkUtil {
+  private static final Logger LOG = LoggerFactory.getLogger(PrometheusMetricsSinkUtil.class);
+
   private static final Pattern SPLIT_PATTERN =
       Pattern.compile("(?<!(^|[A-Z_]))(?=[A-Z])|(?<!^)(?=[A-Z][a-z])");
   private static final Pattern REPLACE_PATTERN =
       Pattern.compile("[^a-zA-Z0-9]+");
 
+  private static final int NORMALIZED_NAME_CACHE_MAX_SIZE = 100_000;
+
+  // Original metric name -> Normalized Prometheus metric name
+  private static final CacheLoader<String, String> NORMALIZED_NAME_CACHE_LOADER =
+      CacheLoader.from(PrometheusMetricsSinkUtil::normalizeImpl);
+  private static final com.google.common.cache.LoadingCache<String, String> NORMALIZED_NAME_CACHE =
+      CacheBuilder.newBuilder()
+          .maximumSize(NORMALIZED_NAME_CACHE_MAX_SIZE)
+          .build(NORMALIZED_NAME_CACHE_LOADER);
+
   /**
    * Never constructed.
    */
@@ -94,11 +112,41 @@ public static String prometheusName(String recordName,
   }
 
   /**
-   * Normalizes metrics tag key name.
-   * @param baseName
-   * @return normalized name.
+   * Normalizes the original metric name to follow the Prometheus naming convention. This method
+   * utilizes cache to improve performance.
+   * <p>
+   *  Reference:
+   *  <ul>
+   *    <li>
+   *      <a href="https://prometheus.io/docs/practices/naming/">
+   *        Metrics and Label Naming</a>
+   *    </li>
+   *    <li>
+   *      <a href="https://prometheus.io/docs/instrumenting/exposition_formats/">
+   *        Exposition formats</a>
+   *    </li>
+   *  </ul>
+   * </p>
+   * @param baseName The original metric name.
+   * @return Prometheus normalized name.
    */
   private static String normalizeName(String baseName) {
+    try {
+      return NORMALIZED_NAME_CACHE.get(baseName);
+    } catch (ExecutionException | UncheckedExecutionException e) {
+      // This should not happen since normalization function do not throw any exception
+      // Nevertheless, we can fall back to uncached implementation if it somehow happens.
+      LOG.warn("Exception encountered when loading metric with base name {} from cache, " +
+          "fall back to uncached normalization implementation", baseName, e);
+      return normalizeImpl(baseName);
+    }
+  }
+
+  /**
+   * Underlying Prometheus normalization implementation.
+   * See {@link PrometheusMetricsSinkUtil#normalizeName(String)} for more information.
+   */
+  private static String normalizeImpl(String baseName) {
     String[] parts = SPLIT_PATTERN.split(baseName);
     String result = String.join("_", parts).toLowerCase();
     return REPLACE_PATTERN.matcher(result).replaceAll("_");