michaeljmarshall commented on code in PR #17241: URL: https://github.com/apache/pulsar/pull/17241#discussion_r966626418
########## managed-ledger/src/main/java/org/apache/bookkeeper/mledger/impl/cache/PendingReadsManager.java: ########## @@ -0,0 +1,447 @@ +/** + * 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.bookkeeper.mledger.impl.cache; + +import static org.apache.bookkeeper.mledger.impl.ManagedLedgerImpl.createManagedLedgerException; +import io.prometheus.client.Counter; +import java.util.ArrayList; +import java.util.List; +import java.util.Map; +import java.util.concurrent.CompletableFuture; +import java.util.concurrent.ConcurrentHashMap; +import java.util.concurrent.atomic.AtomicBoolean; +import lombok.AllArgsConstructor; +import lombok.Value; +import org.apache.bookkeeper.client.api.ReadHandle; +import org.apache.bookkeeper.mledger.AsyncCallbacks; +import org.apache.bookkeeper.mledger.Entry; +import org.apache.bookkeeper.mledger.ManagedLedgerException; +import org.apache.bookkeeper.mledger.impl.EntryImpl; + +/** + * PendingReadsManager tries to prevent sending duplicate reads to BK. + */ +public class PendingReadsManager { + + private static final Counter COUNT_ENTRIES_READ_FROM_BK = Counter + .build() + .name("pulsar_ml_cache_pendingreads_entries_read") + .help("Total number of entries read from BK") + .register(); + + private static final Counter COUNT_ENTRIES_NOTREAD_FROM_BK = Counter + .build() + .name("pulsar_ml_cache_pendingreads_entries_notread") + .help("Total number of entries not read from BK") + .register(); + + private static final Counter COUNT_PENDING_READS_MATCHED = Counter + .build() + .name("pulsar_ml_cache_pendingreads_matched") + .help("Pending reads reused with perfect range match") + .register(); + private static final Counter COUNT_PENDING_READS_MATCHED_INCLUDED = Counter + .build() + .name("pulsar_ml_cache_pendingreads_matched_included") + .help("Pending reads reused by attaching to a read with a larger range") + .register(); + private static final Counter COUNT_PENDING_READS_MISSED = Counter + .build() + .name("pulsar_ml_cache_pendingreads_missed") + .help("Pending reads that didn't find a match") + .register(); + + private static final Counter COUNT_PENDING_READS_MATCHED_OVERLAPPING_MISS_LEFT = Counter + .build() + .name("pulsar_ml_cache_pendingreads_matched_overlapping_miss_left") + .help("Pending reads that didn't find a match but they partially overlap with another read") + .register(); + + private static final Counter COUNT_PENDING_READS_MATCHED_BUT_OVERLAPPING_MISS_RIGHT = Counter + .build() + .name("pulsar_ml_cache_pendingreads_matched_overlapping_miss_right") + .help("Pending reads that didn't find a match but they partially overlap with another read") + .register(); + + private static final Counter COUNT_PENDING_READS_MATCHED_BUT_OVERLAPPING_MISS_BOTH = Counter + .build() + .name("pulsar_ml_cache_pendingreads_matched_overlapping_miss_both") + .help("Pending reads that didn't find a match but they partially overlap with another read") + .register(); Review Comment: I know it is a little late, but I am wondering if there is a reason we created so many metrics here instead of using a single metric with different labels to differentiate certain outcomes. These metrics could probably have been 3 metrics each with 2 or 3 unique labels. For example, `COUNT_PENDING_READS_MATCHED_BUT_OVERLAPPING_MISS_RIGHT` could be `COUNT_PENDING_READS_MATCHED_BUT_OVERLAPPING` with a label indicating that the miss was `right`, `left`, or `both`. In my experience, this makes made Grafana dashboards simpler because you can aggregate the whole metric or break it out by label. I am not aware of any performance implications. Here is the prometheus guide to metric and label names that supports using labels to differentiate a single metric: https://prometheus.io/docs/practices/naming/#labels. -- 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]
