rocksdb/db/merge_helper.h

167 lines
6.7 KiB
C
Raw Normal View History

// Copyright (c) 2013, Facebook, Inc. All rights reserved.
// This source code is licensed under the BSD-style license found in the
// LICENSE file in the root directory of this source tree. An additional grant
// of patent rights can be found in the PATENTS file in the same directory.
//
#ifndef MERGE_HELPER_H
#define MERGE_HELPER_H
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 03:14:32 +00:00
#include <deque>
#include <string>
#include "db/dbformat.h"
#include "rocksdb/compaction_filter.h"
#include "rocksdb/env.h"
#include "rocksdb/slice.h"
#include "util/stop_watch.h"
namespace rocksdb {
class Comparator;
class Iterator;
class Logger;
class MergeOperator;
class Statistics;
class MergeHelper {
public:
MergeHelper(Env* env, const Comparator* user_comparator,
const MergeOperator* user_merge_operator,
const CompactionFilter* compaction_filter, Logger* logger,
unsigned min_partial_merge_operands,
bool assert_valid_internal_key, SequenceNumber latest_snapshot,
int level = 0, Statistics* stats = nullptr)
: env_(env),
user_comparator_(user_comparator),
user_merge_operator_(user_merge_operator),
compaction_filter_(compaction_filter),
logger_(logger),
min_partial_merge_operands_(min_partial_merge_operands),
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 03:14:32 +00:00
assert_valid_internal_key_(assert_valid_internal_key),
latest_snapshot_(latest_snapshot),
level_(level),
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 03:14:32 +00:00
keys_(),
operands_(),
filter_timer_(env_),
total_filter_time_(0U),
stats_(stats) {
assert(user_comparator_ != nullptr);
}
// Wrapper around MergeOperator::FullMerge() that records perf statistics.
// Result of merge will be written to result if status returned is OK.
// If operands is empty, the value will simply be copied to result.
// Returns one of the following statuses:
// - OK: Entries were successfully merged.
// - Corruption: Merge operator reported unsuccessful merge.
// - NotSupported: Merge operator is missing.
static Status TimedFullMerge(const Slice& key, const Slice* value,
const std::deque<std::string>& operands,
const MergeOperator* merge_operator,
Statistics* statistics, Env* env, Logger* logger,
std::string* result);
// Merge entries until we hit
// - a corrupted key
// - a Put/Delete,
// - a different user key,
// - a specific sequence number (snapshot boundary),
// or - the end of iteration
// iter: (IN) points to the first merge type entry
// (OUT) points to the first entry not included in the merge process
// stop_before: (IN) a sequence number that merge should not cross.
// 0 means no restriction
// at_bottom: (IN) true if the iterator covers the bottem level, which means
// we could reach the start of the history of this user key.
//
// Returns one of the following statuses:
// - OK: Entries were successfully merged.
// - MergeInProgress: Put/Delete not encountered and unable to merge operands.
Support for SingleDelete() Summary: This patch fixes #7460559. It introduces SingleDelete as a new database operation. This operation can be used to delete keys that were never overwritten (no put following another put of the same key). If an overwritten key is single deleted the behavior is undefined. Single deletion of a non-existent key has no effect but multiple consecutive single deletions are not allowed (see limitations). In contrast to the conventional Delete() operation, the deletion entry is removed along with the value when the two are lined up in a compaction. Note: The semantics are similar to @igor's prototype that allowed to have this behavior on the granularity of a column family ( https://reviews.facebook.net/D42093 ). This new patch, however, is more aggressive when it comes to removing tombstones: It removes the SingleDelete together with the value whenever there is no snapshot between them while the older patch only did this when the sequence number of the deletion was older than the earliest snapshot. Most of the complex additions are in the Compaction Iterator, all other changes should be relatively straightforward. The patch also includes basic support for single deletions in db_stress and db_bench. Limitations: - Not compatible with cuckoo hash tables - Single deletions cannot be used in combination with merges and normal deletions on the same key (other keys are not affected by this) - Consecutive single deletions are currently not allowed (and older version of this patch supported this so it could be resurrected if needed) Test Plan: make all check Reviewers: yhchiang, sdong, rven, anthony, yoshinorim, igor Reviewed By: igor Subscribers: maykov, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D43179
2015-09-17 18:42:56 +00:00
// - Corruption: Merge operator reported unsuccessful merge or a corrupted
// key has been encountered and not expected (applies only when compiling
// with asserts removed).
//
// REQUIRED: The first key in the input is not corrupted.
Status MergeUntil(Iterator* iter, const SequenceNumber stop_before = 0,
const bool at_bottom = false);
// Filters a merge operand using the compaction filter specified
// in the constructor. Returns true if the operand should be filtered out.
bool FilterMerge(const Slice& user_key, const Slice& value_slice);
// Query the merge result
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 03:14:32 +00:00
// These are valid until the next MergeUntil call
// If the merging was successful:
// - keys() contains a single element with the latest sequence number of
// the merges. The type will be Put or Merge. See IMPORTANT 1 note, below.
// - values() contains a single element with the result of merging all the
// operands together
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 03:14:32 +00:00
//
// IMPORTANT 1: the key type could change after the MergeUntil call.
// Put/Delete + Merge + ... + Merge => Put
// Merge + ... + Merge => Merge
//
// If the merge operator is not associative, and if a Put/Delete is not found
// then the merging will be unsuccessful. In this case:
// - keys() contains the list of internal keys seen in order of iteration.
// - values() contains the list of values (merges) seen in the same order.
// values() is parallel to keys() so that the first entry in
// keys() is the key associated with the first entry in values()
// and so on. These lists will be the same length.
// All of these pairs will be merges over the same user key.
// See IMPORTANT 2 note below.
//
// IMPORTANT 2: The entries were traversed in order from BACK to FRONT.
// So keys().back() was the first key seen by iterator.
// TODO: Re-style this comment to be like the first one
const std::deque<std::string>& keys() const { return keys_; }
const std::deque<std::string>& values() const { return operands_; }
uint64_t TotalFilterTime() const { return total_filter_time_; }
bool HasOperator() const { return user_merge_operator_ != nullptr; }
private:
Env* env_;
const Comparator* user_comparator_;
const MergeOperator* user_merge_operator_;
const CompactionFilter* compaction_filter_;
Logger* logger_;
unsigned min_partial_merge_operands_;
bool assert_valid_internal_key_; // enforce no internal key corruption?
SequenceNumber latest_snapshot_;
int level_;
// the scratch area that holds the result of MergeUntil
// valid up to the next MergeUntil call
[RocksDB] [MergeOperator] The new Merge Interface! Uses merge sequences. Summary: Here are the major changes to the Merge Interface. It has been expanded to handle cases where the MergeOperator is not associative. It does so by stacking up merge operations while scanning through the key history (i.e.: during Get() or Compaction), until a valid Put/Delete/end-of-history is encountered; it then applies all of the merge operations in the correct sequence starting with the base/sentinel value. I have also introduced an "AssociativeMerge" function which allows the user to take advantage of associative merge operations (such as in the case of counters). The implementation will always attempt to merge the operations/operands themselves together when they are encountered, and will resort to the "stacking" method if and only if the "associative-merge" fails. This implementation is conjectured to allow MergeOperator to handle the general case, while still providing the user with the ability to take advantage of certain efficiencies in their own merge-operator / data-structure. NOTE: This is a preliminary diff. This must still go through a lot of review, revision, and testing. Feedback welcome! Test Plan: -This is a preliminary diff. I have only just begun testing/debugging it. -I will be testing this with the existing MergeOperator use-cases and unit-tests (counters, string-append, and redis-lists) -I will be "desk-checking" and walking through the code with the help gdb. -I will find a way of stress-testing the new interface / implementation using db_bench, db_test, merge_test, and/or db_stress. -I will ensure that my tests cover all cases: Get-Memtable, Get-Immutable-Memtable, Get-from-Disk, Iterator-Range-Scan, Flush-Memtable-to-L0, Compaction-L0-L1, Compaction-Ln-L(n+1), Put/Delete found, Put/Delete not-found, end-of-history, end-of-file, etc. -A lot of feedback from the reviewers. Reviewers: haobo, dhruba, zshao, emayanke Reviewed By: haobo CC: leveldb Differential Revision: https://reviews.facebook.net/D11499
2013-08-06 03:14:32 +00:00
std::deque<std::string> keys_; // Keeps track of the sequence of keys seen
std::deque<std::string> operands_; // Parallel with keys_; stores the values
StopWatchNano filter_timer_;
uint64_t total_filter_time_;
Statistics* stats_;
};
// MergeOutputIterator can be used to iterate over the result of a merge.
class MergeOutputIterator {
public:
// The MergeOutputIterator is bound to a MergeHelper instance.
explicit MergeOutputIterator(const MergeHelper* merge_helper);
// Seeks to the first record in the output.
void SeekToFirst();
// Advances to the next record in the output.
void Next();
Slice key() { return Slice(*it_keys_); }
Slice value() { return Slice(*it_values_); }
bool Valid() { return it_keys_ != merge_helper_->keys().rend(); }
private:
const MergeHelper* merge_helper_;
std::deque<std::string>::const_reverse_iterator it_keys_;
std::deque<std::string>::const_reverse_iterator it_values_;
};
} // namespace rocksdb
#endif