path: root/src/mongo/db/prefetch.cpp

/**
 *    Copyright (C) 2018-present MongoDB, Inc.
 *
 *    This program is free software: you can redistribute it and/or modify
 *    it under the terms of the Server Side Public License, version 1,
 *    as published by MongoDB, Inc.
 *
 *    This program is distributed in the hope that it will be useful,
 *    but WITHOUT ANY WARRANTY; without even the implied warranty of
 *    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *    Server Side Public License for more details.
 *
 *    You should have received a copy of the Server Side Public License
 *    along with this program. If not, see
 *    <http://www.mongodb.com/licensing/server-side-public-license>.
 *
 *    As a special exception, the copyright holders give permission to link the
 *    code of portions of this program with the OpenSSL library under certain
 *    conditions as described in each individual source file and distribute
 *    linked combinations including the program with the OpenSSL library. You
 *    must comply with the Server Side Public License in all respects for
 *    all of the code used other than as permitted herein. If you modify file(s)
 *    with this exception, you may extend this exception to your version of the
 *    file(s), but you are not obligated to do so. If you do not wish to do so,
 *    delete this exception statement from your version. If you delete this
 *    exception statement from all source files in the program, then also delete
 *    it in the license file.
 */

#define MONGO_LOG_DEFAULT_COMPONENT ::mongo::logger::LogComponent::kReplication

#include "mongo/platform/basic.h"

#include "mongo/db/prefetch.h"

#include "mongo/db/catalog/collection.h"
#include "mongo/db/catalog/database.h"
#include "mongo/db/catalog/index_catalog.h"
#include "mongo/db/commands/server_status_metric.h"
#include "mongo/db/concurrency/d_concurrency.h"
#include "mongo/db/dbhelpers.h"
#include "mongo/db/index/index_access_method.h"
#include "mongo/db/jsobj.h"
#include "mongo/db/repl/oplog_entry.h"
#include "mongo/db/repl/repl_settings.h"
#include "mongo/db/repl/replication_coordinator.h"
#include "mongo/db/server_parameters.h"
#include "mongo/db/stats/timer_stats.h"
#include "mongo/db/storage/mmap_v1/mmap.h"
#include "mongo/util/log.h"

namespace mongo {

using std::endl;
using std::string;

namespace repl {
namespace {
// todo / idea: the prefetcher, when it fetches _id, on an upsert, will see if the record exists. if
// it does not, at write time, we can just do an insert, which will be faster.

// The count (of batches) and time spent fetching pages before application
//    -- meaning depends on the prefetch behavior: all, _id index, none, etc.)
TimerStats prefetchIndexStats;
ServerStatusMetricField<TimerStats> displayPrefetchIndexPages("repl.preload.indexes",
                                                              &prefetchIndexStats);
TimerStats prefetchDocStats;
ServerStatusMetricField<TimerStats> displayPrefetchDocPages("repl.preload.docs", &prefetchDocStats);

// page in pages needed for all index lookups on a given object
void prefetchIndexPages(OperationContext* opCtx,
                        Collection* collection,
                        const ReplSettings::IndexPrefetchConfig& prefetchConfig,
                        const BSONObj& obj) {
    // do we want prefetchConfig to be (1) as-is, (2) for update ops only, or (3) configured per op
    // type? One might want PREFETCH_NONE for updates, but it's more rare that it is a bad idea for
    // inserts. #3 (per op), a big issue would be "too many knobs".
    switch (prefetchConfig) {
        case ReplSettings::IndexPrefetchConfig::PREFETCH_NONE:
            return;
        case ReplSettings::IndexPrefetchConfig::PREFETCH_ID_ONLY: {
            TimerHolder timer(&prefetchIndexStats);
            // on the update op case, the call to prefetchRecordPages will touch the _id index.
            // thus perhaps this option isn't very useful?
            try {
                IndexDescriptor* desc = collection->getIndexCatalog()->findIdIndex(opCtx);
                if (!desc)
                    return;
                IndexAccessMethod* iam = collection->getIndexCatalog()->getIndex(desc);
                invariant(iam);
                iam->touch(opCtx, obj).transitional_ignore();
            } catch (const DBException& e) {
                LOG(2) << "ignoring exception in prefetchIndexPages(): " << redact(e);
            }
            break;
        }
        case ReplSettings::IndexPrefetchConfig::PREFETCH_ALL: {
            // indexCount includes all indexes, including ones
            // in the process of being built
            IndexCatalog::IndexIterator ii =
                collection->getIndexCatalog()->getIndexIterator(opCtx, true);
            while (ii.more()) {
                TimerHolder timer(&prefetchIndexStats);
                // This will page in all index pages for the given object.
                try {
                    IndexDescriptor* desc = ii.next();
                    IndexAccessMethod* iam = collection->getIndexCatalog()->getIndex(desc);
                    verify(iam);
                    iam->touch(opCtx, obj).transitional_ignore();
                } catch (const DBException& e) {
                    LOG(2) << "ignoring exception in prefetchIndexPages(): " << redact(e);
                }
            }
            break;
        }
        default:
            fassertFailed(16427);
    }
}

// page in the data pages for a record associated with an object
void prefetchRecordPages(OperationContext* opCtx,
                         Database* db,
                         const char* ns,
                         const BSONObj& obj) {
    BSONElement _id;
    if (obj.getObjectID(_id)) {
        TimerHolder timer(&prefetchDocStats);
        BSONObjBuilder builder;
        builder.append(_id);
        BSONObj result;
        try {
            if (Helpers::findById(opCtx, db, ns, builder.done(), result)) {
                // do we want to use Record::touch() here?  it's pretty similar.
                // volatile - avoid compiler optimizations for touching a mmap page
                volatile char _dummy_char = '\0';  // NOLINT

                // Touch the first word on every page in order to fault it into memory
                for (int i = 0; i < result.objsize(); i += getMinOSPageSizeBytes()) {
                    _dummy_char += *(result.objdata() + i);
                }
                // hit the last page, in case we missed it above
                _dummy_char += *(result.objdata() + result.objsize() - 1);
            }
        } catch (const DBException& e) {
            LOG(2) << "ignoring exception in prefetchRecordPages(): " << redact(e);
        }
    }
}
}  // namespace

// prefetch for an oplog operation
void prefetchPagesForReplicatedOp(OperationContext* opCtx,
                                  Database* db,
                                  const OplogEntry& oplogEntry) {
    invariant(db);
    const ReplSettings::IndexPrefetchConfig prefetchConfig =
        ReplicationCoordinator::get(opCtx)->getIndexPrefetchConfig();

    // Prefetch ignores non-CRUD operations.
    if (!oplogEntry.isCrudOpType()) {
        return;
    }

    // This will have to change for engines other than MMAP V1, because they might not have
    // means for directly prefetching pages from the collection. For this purpose, acquire S
    // lock on the database, instead of optimizing with IS.
    const auto& nss = oplogEntry.getNamespace();
    Lock::CollectionLock collLock(opCtx->lockState(), nss.ns(), MODE_S);

    Collection* collection = db->getCollection(opCtx, nss);
    if (!collection) {
        return;
    }

    auto opType = oplogEntry.getOpType();
    LOG(4) << "index prefetch for op " << OpType_serializer(opType);

    // should we prefetch index pages on updates? if the update is in-place and doesn't change
    // indexed values, it is actually slower - a lot slower if there are a dozen indexes or
    // lots of multikeys.  possible variations (not all mutually exclusive):
    //  1) current behavior: full prefetch
    //  2) don't do it for updates
    //  3) don't do multikey indexes for updates
    //  4) don't prefetchIndexPages on some heuristic; e.g., if it's an $inc.
    //  5) if not prefetching index pages (#2), we should do it if we are upsertings and it
    //     will be an insert. to do that we could do the prefetchRecordPage first and if DNE
    //     then we do #1.
    //
    // note that on deletes 'obj' does not have all the keys we would want to prefetch on.
    // a way to achieve that would be to prefetch the record first, and then afterwards do
    // this part.
    //
    auto obj = oplogEntry.getOperationToApply();
    invariant(!obj.isEmpty());
    prefetchIndexPages(opCtx, collection, prefetchConfig, obj);

    // do not prefetch the data for inserts; it doesn't exist yet
    //
    // we should consider doing the record prefetch for the delete op case as we hit the record
    // when we delete.  note if done we only want to touch the first page.
    //
    // update: do record prefetch.
    if ((opType == OpTypeEnum::kUpdate) &&
        // do not prefetch the data for capped collections because
        // they typically do not have an _id index for findById() to use.
        !collection->isCapped()) {
        prefetchRecordPages(opCtx, db, nss.ns().c_str(), obj);
    }
}

class ReplIndexPrefetch : public ServerParameter {
public:
    ReplIndexPrefetch() : ServerParameter(ServerParameterSet::getGlobal(), "replIndexPrefetch") {}

    virtual ~ReplIndexPrefetch() {}

    const char* _value() {
        if (ReplicationCoordinator::get(getGlobalServiceContext())->getReplicationMode() !=
            ReplicationCoordinator::modeReplSet) {
            return "uninitialized";
        }
        ReplSettings::IndexPrefetchConfig ip =
            ReplicationCoordinator::get(getGlobalServiceContext())->getIndexPrefetchConfig();
        switch (ip) {
            case ReplSettings::IndexPrefetchConfig::PREFETCH_NONE:
                return "none";
            case ReplSettings::IndexPrefetchConfig::PREFETCH_ID_ONLY:
                return "_id_only";
            case ReplSettings::IndexPrefetchConfig::PREFETCH_ALL:
                return "all";
            default:
                return "invalid";
        }
    }

    virtual void append(OperationContext* opCtx, BSONObjBuilder& b, const string& name) {
        b.append(name, _value());
    }

    virtual Status set(const BSONElement& newValueElement) {
        if (ReplicationCoordinator::get(getGlobalServiceContext())->getReplicationMode() !=
            ReplicationCoordinator::modeReplSet) {
            return Status(ErrorCodes::BadValue, "replication is not enabled");
        }

        std::string prefetch = newValueElement.valuestrsafe();
        return setFromString(prefetch);
    }

    virtual Status setFromString(const string& prefetch) {
        log() << "changing replication index prefetch behavior to " << prefetch;

        ReplSettings::IndexPrefetchConfig prefetchConfig;

        if (prefetch == "none")
            prefetchConfig = ReplSettings::IndexPrefetchConfig::PREFETCH_NONE;
        else if (prefetch == "_id_only")
            prefetchConfig = ReplSettings::IndexPrefetchConfig::PREFETCH_ID_ONLY;
        else if (prefetch == "all")
            prefetchConfig = ReplSettings::IndexPrefetchConfig::PREFETCH_ALL;
        else {
            return Status(ErrorCodes::BadValue,
                          str::stream() << "unrecognized indexPrefetch setting: " << prefetch);
        }

        ReplicationCoordinator::get(getGlobalServiceContext())
            ->setIndexPrefetchConfig(prefetchConfig);
        return Status::OK();
    }

} replIndexPrefetch;

}  // namespace repl
}  // namespace mongo