You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
rocksdb/table/meta_blocks.cc

525 lines
20 KiB

// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
// This source code is licensed under both the GPLv2 (found in the
// COPYING file in the root directory) and Apache 2.0 License
// (found in the LICENSE.Apache file in the root directory).
#include "table/meta_blocks.h"
#include <map>
#include <string>
#include "block_fetcher.h"
#include "db/table_properties_collector.h"
#include "rocksdb/table.h"
#include "rocksdb/table_properties.h"
#include "table/block_based/block.h"
#include "table/format.h"
#include "table/internal_iterator.h"
#include "table/persistent_cache_helper.h"
#include "table/table_properties_internal.h"
#include "test_util/sync_point.h"
#include "util/coding.h"
#include "util/file_reader_writer.h"
namespace rocksdb {
MetaIndexBuilder::MetaIndexBuilder()
: meta_index_block_(new BlockBuilder(1 /* restart interval */)) {}
void MetaIndexBuilder::Add(const std::string& key,
const BlockHandle& handle) {
std::string handle_encoding;
handle.EncodeTo(&handle_encoding);
meta_block_handles_.insert({key, handle_encoding});
}
Slice MetaIndexBuilder::Finish() {
for (const auto& metablock : meta_block_handles_) {
meta_index_block_->Add(metablock.first, metablock.second);
}
return meta_index_block_->Finish();
}
// Property block will be read sequentially and cached in a heap located
// object, so there's no need for restart points. Thus we set the restart
// interval to infinity to save space.
PropertyBlockBuilder::PropertyBlockBuilder()
: properties_block_(
new BlockBuilder(port::kMaxInt32 /* restart interval */)) {}
void PropertyBlockBuilder::Add(const std::string& name,
const std::string& val) {
props_.insert({name, val});
}
void PropertyBlockBuilder::Add(const std::string& name, uint64_t val) {
assert(props_.find(name) == props_.end());
std::string dst;
PutVarint64(&dst, val);
Add(name, dst);
}
void PropertyBlockBuilder::Add(
const UserCollectedProperties& user_collected_properties) {
for (const auto& prop : user_collected_properties) {
Add(prop.first, prop.second);
}
}
void PropertyBlockBuilder::AddTableProperty(const TableProperties& props) {
TEST_SYNC_POINT_CALLBACK("PropertyBlockBuilder::AddTableProperty:Start",
const_cast<TableProperties*>(&props));
Add(TablePropertiesNames::kRawKeySize, props.raw_key_size);
Add(TablePropertiesNames::kRawValueSize, props.raw_value_size);
Add(TablePropertiesNames::kDataSize, props.data_size);
Add(TablePropertiesNames::kIndexSize, props.index_size);
if (props.index_partitions != 0) {
Add(TablePropertiesNames::kIndexPartitions, props.index_partitions);
Add(TablePropertiesNames::kTopLevelIndexSize, props.top_level_index_size);
}
Add(TablePropertiesNames::kIndexKeyIsUserKey, props.index_key_is_user_key);
Add(TablePropertiesNames::kIndexValueIsDeltaEncoded,
props.index_value_is_delta_encoded);
Add(TablePropertiesNames::kNumEntries, props.num_entries);
Add(TablePropertiesNames::kDeletedKeys, props.num_deletions);
Add(TablePropertiesNames::kMergeOperands, props.num_merge_operands);
Add(TablePropertiesNames::kNumRangeDeletions, props.num_range_deletions);
Add(TablePropertiesNames::kNumDataBlocks, props.num_data_blocks);
Add(TablePropertiesNames::kFilterSize, props.filter_size);
Add(TablePropertiesNames::kFormatVersion, props.format_version);
Add(TablePropertiesNames::kFixedKeyLen, props.fixed_key_len);
Add(TablePropertiesNames::kColumnFamilyId, props.column_family_id);
FIFO Compaction with TTL Summary: Introducing FIFO compactions with TTL. FIFO compaction is based on size only which makes it tricky to enable in production as use cases can have organic growth. A user requested an option to drop files based on the time of their creation instead of the total size. To address that request: - Added a new TTL option to FIFO compaction options. - Updated FIFO compaction score to take TTL into consideration. - Added a new table property, creation_time, to keep track of when the SST file is created. - Creation_time is set as below: - On Flush: Set to the time of flush. - On Compaction: Set to the max creation_time of all the files involved in the compaction. - On Repair and Recovery: Set to the time of repair/recovery. - Old files created prior to this code change will have a creation_time of 0. - FIFO compaction with TTL is enabled when ttl > 0. All files older than ttl will be deleted during compaction. i.e. `if (file.creation_time < (current_time - ttl)) then delete(file)`. This will enable cases where you might want to delete all files older than, say, 1 day. - FIFO compaction will fall back to the prior way of deleting files based on size if: - the creation_time of all files involved in compaction is 0. - the total size (of all SST files combined) does not drop below `compaction_options_fifo.max_table_files_size` even if the files older than ttl are deleted. This feature is not supported if max_open_files != -1 or with table formats other than Block-based. **Test Plan:** Added tests. **Benchmark results:** Base: FIFO with max size: 100MB :: ``` svemuri@dev15905 ~/rocksdb (fifo-compaction) $ TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=readwhilewriting --num=5000000 --threads=16 --compaction_style=2 --fifo_compaction_max_table_files_size_mb=100 readwhilewriting : 1.924 micros/op 519858 ops/sec; 13.6 MB/s (1176277 of 5000000 found) ``` With TTL (a low one for testing) :: ``` svemuri@dev15905 ~/rocksdb (fifo-compaction) $ TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=readwhilewriting --num=5000000 --threads=16 --compaction_style=2 --fifo_compaction_max_table_files_size_mb=100 --fifo_compaction_ttl=20 readwhilewriting : 1.902 micros/op 525817 ops/sec; 13.7 MB/s (1185057 of 5000000 found) ``` Example Log lines: ``` 2017/06/26-15:17:24.609249 7fd5a45ff700 (Original Log Time 2017/06/26-15:17:24.609177) [db/compaction_picker.cc:1471] [default] FIFO compaction: picking file 40 with creation time 1498515423 for deletion 2017/06/26-15:17:24.609255 7fd5a45ff700 (Original Log Time 2017/06/26-15:17:24.609234) [db/db_impl_compaction_flush.cc:1541] [default] Deleted 1 files ... 2017/06/26-15:17:25.553185 7fd5a61a5800 [DEBUG] [db/db_impl_files.cc:309] [JOB 0] Delete /dev/shm/dbbench/000040.sst type=2 #40 -- OK 2017/06/26-15:17:25.553205 7fd5a61a5800 EVENT_LOG_v1 {"time_micros": 1498515445553199, "job": 0, "event": "table_file_deletion", "file_number": 40} ``` SST Files remaining in the dbbench dir, after db_bench execution completed: ``` svemuri@dev15905 ~/rocksdb (fifo-compaction) $ ls -l /dev/shm//dbbench/*.sst -rw-r--r--. 1 svemuri users 30749887 Jun 26 15:17 /dev/shm//dbbench/000042.sst -rw-r--r--. 1 svemuri users 30768779 Jun 26 15:17 /dev/shm//dbbench/000044.sst -rw-r--r--. 1 svemuri users 30757481 Jun 26 15:17 /dev/shm//dbbench/000046.sst ``` Closes https://github.com/facebook/rocksdb/pull/2480 Differential Revision: D5305116 Pulled By: sagar0 fbshipit-source-id: 3e5cfcf5dd07ed2211b5b37492eb235b45139174
8 years ago
Add(TablePropertiesNames::kCreationTime, props.creation_time);
Add(TablePropertiesNames::kOldestKeyTime, props.oldest_key_time);
Periodic Compactions (#5166) Summary: Introducing Periodic Compactions. This feature allows all the files in a CF to be periodically compacted. It could help in catching any corruptions that could creep into the DB proactively as every file is constantly getting re-compacted. And also, of course, it helps to cleanup data older than certain threshold. - Introduced a new option `periodic_compaction_time` to control how long a file can live without being compacted in a CF. - This works across all levels. - The files are put in the same level after going through the compaction. (Related files in the same level are picked up as `ExpandInputstoCleanCut` is used). - Compaction filters, if any, are invoked as usual. - A new table property, `file_creation_time`, is introduced to implement this feature. This property is set to the time at which the SST file was created (and that time is given by the underlying Env/OS). This feature can be enabled on its own, or in conjunction with `ttl`. It is possible to set a different time threshold for the bottom level when used in conjunction with ttl. Since `ttl` works only on 0 to last but one levels, you could set `ttl` to, say, 1 day, and `periodic_compaction_time` to, say, 7 days. Since `ttl < periodic_compaction_time` all files in last but one levels keep getting picked up based on ttl, and almost never based on periodic_compaction_time. The files in the bottom level get picked up for compaction based on `periodic_compaction_time`. Pull Request resolved: https://github.com/facebook/rocksdb/pull/5166 Differential Revision: D14884441 Pulled By: sagar0 fbshipit-source-id: 408426cbacb409c06386a98632dcf90bfa1bda47
6 years ago
if (props.file_creation_time > 0) {
Add(TablePropertiesNames::kFileCreationTime, props.file_creation_time);
}
if (!props.filter_policy_name.empty()) {
Add(TablePropertiesNames::kFilterPolicy, props.filter_policy_name);
}
if (!props.comparator_name.empty()) {
Add(TablePropertiesNames::kComparator, props.comparator_name);
}
if (!props.merge_operator_name.empty()) {
Add(TablePropertiesNames::kMergeOperator, props.merge_operator_name);
}
store prefix_extractor_name in table Summary: Make sure prefix extractor name is stored in SST files and if DB is opened with a prefix extractor of a different name, prefix bloom is skipped when read the file. Also add unit tests for that. Test Plan: before change: ``` Note: Google Test filter = BlockBasedTableTest.SkipPrefixBloomFilter [==========] Running 1 test from 1 test case. [----------] Global test environment set-up. [----------] 1 test from BlockBasedTableTest [ RUN ] BlockBasedTableTest.SkipPrefixBloomFilter table/table_test.cc:1421: Failure Value of: db_iter->Valid() Actual: false Expected: true [ FAILED ] BlockBasedTableTest.SkipPrefixBloomFilter (1 ms) [----------] 1 test from BlockBasedTableTest (1 ms total) [----------] Global test environment tear-down [==========] 1 test from 1 test case ran. (1 ms total) [ PASSED ] 0 tests. [ FAILED ] 1 test, listed below: [ FAILED ] BlockBasedTableTest.SkipPrefixBloomFilter 1 FAILED TEST ``` after: ``` Note: Google Test filter = BlockBasedTableTest.SkipPrefixBloomFilter [==========] Running 1 test from 1 test case. [----------] Global test environment set-up. [----------] 1 test from BlockBasedTableTest [ RUN ] BlockBasedTableTest.SkipPrefixBloomFilter [ OK ] BlockBasedTableTest.SkipPrefixBloomFilter (0 ms) [----------] 1 test from BlockBasedTableTest (0 ms total) [----------] Global test environment tear-down [==========] 1 test from 1 test case ran. (0 ms total) [ PASSED ] 1 test. ``` Reviewers: sdong, andrewkr, yiwu, IslamAbdelRahman Reviewed By: IslamAbdelRahman Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D61215
8 years ago
if (!props.prefix_extractor_name.empty()) {
Add(TablePropertiesNames::kPrefixExtractorName,
props.prefix_extractor_name);
}
if (!props.property_collectors_names.empty()) {
Add(TablePropertiesNames::kPropertyCollectors,
props.property_collectors_names);
}
if (!props.column_family_name.empty()) {
Add(TablePropertiesNames::kColumnFamilyName, props.column_family_name);
}
if (!props.compression_name.empty()) {
Add(TablePropertiesNames::kCompression, props.compression_name);
}
add compression options to table properties (#5081) Summary: Since we are planning to use dictionary compression and to use different compression level, it is quite useful to add compression options to TableProperties. For example, in MyRocks, if the feature is available, we can query from information_schema.rocksdb_sst_props to see if all sst files are converted to ZSTD dictionary compressions. Resolves https://github.com/facebook/rocksdb/issues/4992 With this PR, user can query table properties through `GetPropertiesOfAllTables` API and get compression options as std::string: `window_bits=-14; level=32767; strategy=0; max_dict_bytes=0; zstd_max_train_bytes=0; enabled=0;` or table_properties->ToString() will also contain it `# data blocks=1; # entries=13; # deletions=0; # merge operands=0; # range deletions=0; raw key size=143; raw average key size=11.000000; raw value size=39; raw average value size=3.000000; data block size=120; index block size (user-key? 0, delta-value? 0)=27; filter block size=0; (estimated) table size=147; filter policy name=N/A; prefix extractor name=nullptr; column family ID=0; column family name=default; comparator name=leveldb.BytewiseComparator; merge operator name=nullptr; property collectors names=[]; SST file compression algo=Snappy; SST file compression options=window_bits=-14; level=32767; strategy=0; max_dict_bytes=0; zstd_max_train_bytes=0; enabled=0; ; creation time=1552946632; time stamp of earliest key=1552946632;` Pull Request resolved: https://github.com/facebook/rocksdb/pull/5081 Differential Revision: D14716692 Pulled By: miasantreble fbshipit-source-id: 7d2f2cf84e052bff876e71b4212cfdebf5be32dd
6 years ago
if (!props.compression_options.empty()) {
Add(TablePropertiesNames::kCompressionOptions, props.compression_options);
}
}
Slice PropertyBlockBuilder::Finish() {
for (const auto& prop : props_) {
properties_block_->Add(prop.first, prop.second);
}
return properties_block_->Finish();
}
void LogPropertiesCollectionError(
Logger* info_log, const std::string& method, const std::string& name) {
assert(method == "Add" || method == "Finish");
std::string msg =
"Encountered error when calling TablePropertiesCollector::" +
method + "() with collector name: " + name;
ROCKS_LOG_ERROR(info_log, "%s", msg.c_str());
}
bool NotifyCollectTableCollectorsOnAdd(
const Slice& key, const Slice& value, uint64_t file_size,
const std::vector<std::unique_ptr<IntTblPropCollector>>& collectors,
Logger* info_log) {
bool all_succeeded = true;
TablePropertiesCollectorFactory Summary: This diff addresses task #4296714 and rethinks how users provide us with TablePropertiesCollectors as part of Options. Here's description of task #4296714: I'm debugging #4295529 and noticed that our count of user properties kDeletedKeys is wrong. We're sharing one single InternalKeyPropertiesCollector with all Table Builders. In LOG Files, we're outputting number of kDeletedKeys as connected with a single table, while it's actually the total count of deleted keys since creation of the DB. For example, this table has 3155 entries and 1391828 deleted keys. The problem with current approach that we call methods on a single TablePropertiesCollector for all the tables we create. Even worse, we could do it from multiple threads at the same time and TablePropertiesCollector has no way of knowing which table we're calling it for. Good part: Looks like nobody inside Facebook is using Options::table_properties_collectors. This means we should be able to painfully change the API. In this change, I introduce TablePropertiesCollectorFactory. For every table we create, we call `CreateTablePropertiesCollector`, which creates a TablePropertiesCollector for a single table. We then use it sequentially from a single thread, which means it doesn't have to be thread-safe. Test Plan: Added a test in table_properties_collector_test that fails on master (build two tables, assert that kDeletedKeys count is correct for the second one). Also, all other tests Reviewers: sdong, dhruba, haobo, kailiu Reviewed By: kailiu CC: leveldb Differential Revision: https://reviews.facebook.net/D18579
11 years ago
for (auto& collector : collectors) {
Status s = collector->InternalAdd(key, value, file_size);
all_succeeded = all_succeeded && s.ok();
if (!s.ok()) {
LogPropertiesCollectionError(info_log, "Add" /* method */,
collector->Name());
}
}
return all_succeeded;
}
void NotifyCollectTableCollectorsOnBlockAdd(
const std::vector<std::unique_ptr<IntTblPropCollector>>& collectors,
const uint64_t blockRawBytes, const uint64_t blockCompressedBytesFast,
const uint64_t blockCompressedBytesSlow) {
for (auto& collector : collectors) {
collector->BlockAdd(blockRawBytes, blockCompressedBytesFast,
blockCompressedBytesSlow);
}
}
bool NotifyCollectTableCollectorsOnFinish(
const std::vector<std::unique_ptr<IntTblPropCollector>>& collectors,
TablePropertiesCollectorFactory Summary: This diff addresses task #4296714 and rethinks how users provide us with TablePropertiesCollectors as part of Options. Here's description of task #4296714: I'm debugging #4295529 and noticed that our count of user properties kDeletedKeys is wrong. We're sharing one single InternalKeyPropertiesCollector with all Table Builders. In LOG Files, we're outputting number of kDeletedKeys as connected with a single table, while it's actually the total count of deleted keys since creation of the DB. For example, this table has 3155 entries and 1391828 deleted keys. The problem with current approach that we call methods on a single TablePropertiesCollector for all the tables we create. Even worse, we could do it from multiple threads at the same time and TablePropertiesCollector has no way of knowing which table we're calling it for. Good part: Looks like nobody inside Facebook is using Options::table_properties_collectors. This means we should be able to painfully change the API. In this change, I introduce TablePropertiesCollectorFactory. For every table we create, we call `CreateTablePropertiesCollector`, which creates a TablePropertiesCollector for a single table. We then use it sequentially from a single thread, which means it doesn't have to be thread-safe. Test Plan: Added a test in table_properties_collector_test that fails on master (build two tables, assert that kDeletedKeys count is correct for the second one). Also, all other tests Reviewers: sdong, dhruba, haobo, kailiu Reviewed By: kailiu CC: leveldb Differential Revision: https://reviews.facebook.net/D18579
11 years ago
Logger* info_log, PropertyBlockBuilder* builder) {
bool all_succeeded = true;
TablePropertiesCollectorFactory Summary: This diff addresses task #4296714 and rethinks how users provide us with TablePropertiesCollectors as part of Options. Here's description of task #4296714: I'm debugging #4295529 and noticed that our count of user properties kDeletedKeys is wrong. We're sharing one single InternalKeyPropertiesCollector with all Table Builders. In LOG Files, we're outputting number of kDeletedKeys as connected with a single table, while it's actually the total count of deleted keys since creation of the DB. For example, this table has 3155 entries and 1391828 deleted keys. The problem with current approach that we call methods on a single TablePropertiesCollector for all the tables we create. Even worse, we could do it from multiple threads at the same time and TablePropertiesCollector has no way of knowing which table we're calling it for. Good part: Looks like nobody inside Facebook is using Options::table_properties_collectors. This means we should be able to painfully change the API. In this change, I introduce TablePropertiesCollectorFactory. For every table we create, we call `CreateTablePropertiesCollector`, which creates a TablePropertiesCollector for a single table. We then use it sequentially from a single thread, which means it doesn't have to be thread-safe. Test Plan: Added a test in table_properties_collector_test that fails on master (build two tables, assert that kDeletedKeys count is correct for the second one). Also, all other tests Reviewers: sdong, dhruba, haobo, kailiu Reviewed By: kailiu CC: leveldb Differential Revision: https://reviews.facebook.net/D18579
11 years ago
for (auto& collector : collectors) {
UserCollectedProperties user_collected_properties;
Status s = collector->Finish(&user_collected_properties);
all_succeeded = all_succeeded && s.ok();
if (!s.ok()) {
LogPropertiesCollectionError(info_log, "Finish" /* method */,
collector->Name());
} else {
builder->Add(user_collected_properties);
}
}
return all_succeeded;
}
Status ReadProperties(const Slice& handle_value, RandomAccessFileReader* file,
FilePrefetchBuffer* prefetch_buffer, const Footer& footer,
const ImmutableCFOptions& ioptions,
TableProperties** table_properties, bool verify_checksum,
BlockHandle* ret_block_handle,
CacheAllocationPtr* verification_buf,
bool /*compression_type_missing*/,
MemoryAllocator* memory_allocator) {
assert(table_properties);
Slice v = handle_value;
BlockHandle handle;
if (!handle.DecodeFrom(&v).ok()) {
return Status::InvalidArgument("Failed to decode properties block handle");
}
BlockContents block_contents;
ReadOptions read_options;
read_options.verify_checksums = verify_checksum;
Status s;
PersistentCacheOptions cache_options;
BlockFetcher block_fetcher(
file, prefetch_buffer, footer, read_options, handle, &block_contents,
ioptions, false /* decompress */, false /*maybe_compressed*/,
BlockType::kProperties, UncompressionDict::GetEmptyDict(), cache_options,
memory_allocator);
s = block_fetcher.ReadBlockContents();
// property block is never compressed. Need to add uncompress logic if we are
// to compress it..
if (!s.ok()) {
return s;
}
Block properties_block(std::move(block_contents),
kDisableGlobalSequenceNumber);
DataBlockIter iter;
Add an option to put first key of each sst block in the index (#5289) Summary: The first key is used to defer reading the data block until this file gets to the top of merging iterator's heap. For short range scans, most files never make it to the top of the heap, so this change can reduce read amplification by a lot sometimes. Consider the following workload. There are a few data streams (we'll be calling them "logs"), each stream consisting of a sequence of blobs (we'll be calling them "records"). Each record is identified by log ID and a sequence number within the log. RocksDB key is concatenation of log ID and sequence number (big endian). Reads are mostly relatively short range scans, each within a single log. Writes are mostly sequential for each log, but writes to different logs are randomly interleaved. Compactions are disabled; instead, when we accumulate a few tens of sst files, we create a new column family and start writing to it. So, a typical sst file consists of a few ranges of blocks, each range corresponding to one log ID (we use FlushBlockPolicy to cut blocks at log boundaries). A typical read would go like this. First, iterator Seek() reads one block from each sst file. Then a series of Next()s move through one sst file (since writes to each log are mostly sequential) until the subiterator reaches the end of this log in this sst file; then Next() switches to the next sst file and reads sequentially from that, and so on. Often a range scan will only return records from a small number of blocks in small number of sst files; in this case, the cost of initial Seek() reading one block from each file may be bigger than the cost of reading the actually useful blocks. Neither iterate_upper_bound nor bloom filters can prevent reading one block from each file in Seek(). But this PR can: if the index contains first key from each block, we don't have to read the block until this block actually makes it to the top of merging iterator's heap, so for short range scans we won't read any blocks from most of the sst files. This PR does the deferred block loading inside value() call. This is not ideal: there's no good way to report an IO error from inside value(). As discussed with siying offline, it would probably be better to change InternalIterator's interface to explicitly fetch deferred value and get status. I'll do it in a separate PR. Pull Request resolved: https://github.com/facebook/rocksdb/pull/5289 Differential Revision: D15256423 Pulled By: al13n321 fbshipit-source-id: 750e4c39ce88e8d41662f701cf6275d9388ba46a
6 years ago
properties_block.NewDataIterator(BytewiseComparator(), BytewiseComparator(),
&iter);
auto new_table_properties = new TableProperties();
// All pre-defined properties of type uint64_t
std::unordered_map<std::string, uint64_t*> predefined_uint64_properties = {
{TablePropertiesNames::kDataSize, &new_table_properties->data_size},
{TablePropertiesNames::kIndexSize, &new_table_properties->index_size},
{TablePropertiesNames::kIndexPartitions,
&new_table_properties->index_partitions},
{TablePropertiesNames::kTopLevelIndexSize,
&new_table_properties->top_level_index_size},
{TablePropertiesNames::kIndexKeyIsUserKey,
&new_table_properties->index_key_is_user_key},
{TablePropertiesNames::kIndexValueIsDeltaEncoded,
&new_table_properties->index_value_is_delta_encoded},
{TablePropertiesNames::kFilterSize, &new_table_properties->filter_size},
{TablePropertiesNames::kRawKeySize, &new_table_properties->raw_key_size},
{TablePropertiesNames::kRawValueSize,
&new_table_properties->raw_value_size},
{TablePropertiesNames::kNumDataBlocks,
&new_table_properties->num_data_blocks},
{TablePropertiesNames::kNumEntries, &new_table_properties->num_entries},
{TablePropertiesNames::kDeletedKeys,
&new_table_properties->num_deletions},
{TablePropertiesNames::kMergeOperands,
&new_table_properties->num_merge_operands},
{TablePropertiesNames::kNumRangeDeletions,
&new_table_properties->num_range_deletions},
{TablePropertiesNames::kFormatVersion,
&new_table_properties->format_version},
{TablePropertiesNames::kFixedKeyLen,
&new_table_properties->fixed_key_len},
{TablePropertiesNames::kColumnFamilyId,
&new_table_properties->column_family_id},
FIFO Compaction with TTL Summary: Introducing FIFO compactions with TTL. FIFO compaction is based on size only which makes it tricky to enable in production as use cases can have organic growth. A user requested an option to drop files based on the time of their creation instead of the total size. To address that request: - Added a new TTL option to FIFO compaction options. - Updated FIFO compaction score to take TTL into consideration. - Added a new table property, creation_time, to keep track of when the SST file is created. - Creation_time is set as below: - On Flush: Set to the time of flush. - On Compaction: Set to the max creation_time of all the files involved in the compaction. - On Repair and Recovery: Set to the time of repair/recovery. - Old files created prior to this code change will have a creation_time of 0. - FIFO compaction with TTL is enabled when ttl > 0. All files older than ttl will be deleted during compaction. i.e. `if (file.creation_time < (current_time - ttl)) then delete(file)`. This will enable cases where you might want to delete all files older than, say, 1 day. - FIFO compaction will fall back to the prior way of deleting files based on size if: - the creation_time of all files involved in compaction is 0. - the total size (of all SST files combined) does not drop below `compaction_options_fifo.max_table_files_size` even if the files older than ttl are deleted. This feature is not supported if max_open_files != -1 or with table formats other than Block-based. **Test Plan:** Added tests. **Benchmark results:** Base: FIFO with max size: 100MB :: ``` svemuri@dev15905 ~/rocksdb (fifo-compaction) $ TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=readwhilewriting --num=5000000 --threads=16 --compaction_style=2 --fifo_compaction_max_table_files_size_mb=100 readwhilewriting : 1.924 micros/op 519858 ops/sec; 13.6 MB/s (1176277 of 5000000 found) ``` With TTL (a low one for testing) :: ``` svemuri@dev15905 ~/rocksdb (fifo-compaction) $ TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=readwhilewriting --num=5000000 --threads=16 --compaction_style=2 --fifo_compaction_max_table_files_size_mb=100 --fifo_compaction_ttl=20 readwhilewriting : 1.902 micros/op 525817 ops/sec; 13.7 MB/s (1185057 of 5000000 found) ``` Example Log lines: ``` 2017/06/26-15:17:24.609249 7fd5a45ff700 (Original Log Time 2017/06/26-15:17:24.609177) [db/compaction_picker.cc:1471] [default] FIFO compaction: picking file 40 with creation time 1498515423 for deletion 2017/06/26-15:17:24.609255 7fd5a45ff700 (Original Log Time 2017/06/26-15:17:24.609234) [db/db_impl_compaction_flush.cc:1541] [default] Deleted 1 files ... 2017/06/26-15:17:25.553185 7fd5a61a5800 [DEBUG] [db/db_impl_files.cc:309] [JOB 0] Delete /dev/shm/dbbench/000040.sst type=2 #40 -- OK 2017/06/26-15:17:25.553205 7fd5a61a5800 EVENT_LOG_v1 {"time_micros": 1498515445553199, "job": 0, "event": "table_file_deletion", "file_number": 40} ``` SST Files remaining in the dbbench dir, after db_bench execution completed: ``` svemuri@dev15905 ~/rocksdb (fifo-compaction) $ ls -l /dev/shm//dbbench/*.sst -rw-r--r--. 1 svemuri users 30749887 Jun 26 15:17 /dev/shm//dbbench/000042.sst -rw-r--r--. 1 svemuri users 30768779 Jun 26 15:17 /dev/shm//dbbench/000044.sst -rw-r--r--. 1 svemuri users 30757481 Jun 26 15:17 /dev/shm//dbbench/000046.sst ``` Closes https://github.com/facebook/rocksdb/pull/2480 Differential Revision: D5305116 Pulled By: sagar0 fbshipit-source-id: 3e5cfcf5dd07ed2211b5b37492eb235b45139174
8 years ago
{TablePropertiesNames::kCreationTime,
&new_table_properties->creation_time},
{TablePropertiesNames::kOldestKeyTime,
&new_table_properties->oldest_key_time},
Periodic Compactions (#5166) Summary: Introducing Periodic Compactions. This feature allows all the files in a CF to be periodically compacted. It could help in catching any corruptions that could creep into the DB proactively as every file is constantly getting re-compacted. And also, of course, it helps to cleanup data older than certain threshold. - Introduced a new option `periodic_compaction_time` to control how long a file can live without being compacted in a CF. - This works across all levels. - The files are put in the same level after going through the compaction. (Related files in the same level are picked up as `ExpandInputstoCleanCut` is used). - Compaction filters, if any, are invoked as usual. - A new table property, `file_creation_time`, is introduced to implement this feature. This property is set to the time at which the SST file was created (and that time is given by the underlying Env/OS). This feature can be enabled on its own, or in conjunction with `ttl`. It is possible to set a different time threshold for the bottom level when used in conjunction with ttl. Since `ttl` works only on 0 to last but one levels, you could set `ttl` to, say, 1 day, and `periodic_compaction_time` to, say, 7 days. Since `ttl < periodic_compaction_time` all files in last but one levels keep getting picked up based on ttl, and almost never based on periodic_compaction_time. The files in the bottom level get picked up for compaction based on `periodic_compaction_time`. Pull Request resolved: https://github.com/facebook/rocksdb/pull/5166 Differential Revision: D14884441 Pulled By: sagar0 fbshipit-source-id: 408426cbacb409c06386a98632dcf90bfa1bda47
6 years ago
{TablePropertiesNames::kFileCreationTime,
&new_table_properties->file_creation_time},
};
std::string last_key;
for (iter.SeekToFirstOrReport(); iter.Valid(); iter.NextOrReport()) {
s = iter.status();
if (!s.ok()) {
break;
}
auto key = iter.key().ToString();
// properties block should be strictly sorted with no duplicate key.
if (!last_key.empty() &&
BytewiseComparator()->Compare(key, last_key) <= 0) {
s = Status::Corruption("properties unsorted");
break;
}
last_key = key;
auto raw_val = iter.value();
auto pos = predefined_uint64_properties.find(key);
new_table_properties->properties_offsets.insert(
{key, handle.offset() + iter.ValueOffset()});
if (pos != predefined_uint64_properties.end()) {
if (key == TablePropertiesNames::kDeletedKeys ||
key == TablePropertiesNames::kMergeOperands) {
// Insert in user-collected properties for API backwards compatibility
new_table_properties->user_collected_properties.insert(
{key, raw_val.ToString()});
}
// handle predefined rocksdb properties
uint64_t val;
if (!GetVarint64(&raw_val, &val)) {
// skip malformed value
auto error_msg =
"Detect malformed value in properties meta-block:"
"\tkey: " + key + "\tval: " + raw_val.ToString();
ROCKS_LOG_ERROR(ioptions.info_log, "%s", error_msg.c_str());
continue;
}
*(pos->second) = val;
} else if (key == TablePropertiesNames::kFilterPolicy) {
new_table_properties->filter_policy_name = raw_val.ToString();
} else if (key == TablePropertiesNames::kColumnFamilyName) {
new_table_properties->column_family_name = raw_val.ToString();
} else if (key == TablePropertiesNames::kComparator) {
new_table_properties->comparator_name = raw_val.ToString();
} else if (key == TablePropertiesNames::kMergeOperator) {
new_table_properties->merge_operator_name = raw_val.ToString();
store prefix_extractor_name in table Summary: Make sure prefix extractor name is stored in SST files and if DB is opened with a prefix extractor of a different name, prefix bloom is skipped when read the file. Also add unit tests for that. Test Plan: before change: ``` Note: Google Test filter = BlockBasedTableTest.SkipPrefixBloomFilter [==========] Running 1 test from 1 test case. [----------] Global test environment set-up. [----------] 1 test from BlockBasedTableTest [ RUN ] BlockBasedTableTest.SkipPrefixBloomFilter table/table_test.cc:1421: Failure Value of: db_iter->Valid() Actual: false Expected: true [ FAILED ] BlockBasedTableTest.SkipPrefixBloomFilter (1 ms) [----------] 1 test from BlockBasedTableTest (1 ms total) [----------] Global test environment tear-down [==========] 1 test from 1 test case ran. (1 ms total) [ PASSED ] 0 tests. [ FAILED ] 1 test, listed below: [ FAILED ] BlockBasedTableTest.SkipPrefixBloomFilter 1 FAILED TEST ``` after: ``` Note: Google Test filter = BlockBasedTableTest.SkipPrefixBloomFilter [==========] Running 1 test from 1 test case. [----------] Global test environment set-up. [----------] 1 test from BlockBasedTableTest [ RUN ] BlockBasedTableTest.SkipPrefixBloomFilter [ OK ] BlockBasedTableTest.SkipPrefixBloomFilter (0 ms) [----------] 1 test from BlockBasedTableTest (0 ms total) [----------] Global test environment tear-down [==========] 1 test from 1 test case ran. (0 ms total) [ PASSED ] 1 test. ``` Reviewers: sdong, andrewkr, yiwu, IslamAbdelRahman Reviewed By: IslamAbdelRahman Subscribers: andrewkr, dhruba, leveldb Differential Revision: https://reviews.facebook.net/D61215
8 years ago
} else if (key == TablePropertiesNames::kPrefixExtractorName) {
new_table_properties->prefix_extractor_name = raw_val.ToString();
} else if (key == TablePropertiesNames::kPropertyCollectors) {
new_table_properties->property_collectors_names = raw_val.ToString();
} else if (key == TablePropertiesNames::kCompression) {
new_table_properties->compression_name = raw_val.ToString();
add compression options to table properties (#5081) Summary: Since we are planning to use dictionary compression and to use different compression level, it is quite useful to add compression options to TableProperties. For example, in MyRocks, if the feature is available, we can query from information_schema.rocksdb_sst_props to see if all sst files are converted to ZSTD dictionary compressions. Resolves https://github.com/facebook/rocksdb/issues/4992 With this PR, user can query table properties through `GetPropertiesOfAllTables` API and get compression options as std::string: `window_bits=-14; level=32767; strategy=0; max_dict_bytes=0; zstd_max_train_bytes=0; enabled=0;` or table_properties->ToString() will also contain it `# data blocks=1; # entries=13; # deletions=0; # merge operands=0; # range deletions=0; raw key size=143; raw average key size=11.000000; raw value size=39; raw average value size=3.000000; data block size=120; index block size (user-key? 0, delta-value? 0)=27; filter block size=0; (estimated) table size=147; filter policy name=N/A; prefix extractor name=nullptr; column family ID=0; column family name=default; comparator name=leveldb.BytewiseComparator; merge operator name=nullptr; property collectors names=[]; SST file compression algo=Snappy; SST file compression options=window_bits=-14; level=32767; strategy=0; max_dict_bytes=0; zstd_max_train_bytes=0; enabled=0; ; creation time=1552946632; time stamp of earliest key=1552946632;` Pull Request resolved: https://github.com/facebook/rocksdb/pull/5081 Differential Revision: D14716692 Pulled By: miasantreble fbshipit-source-id: 7d2f2cf84e052bff876e71b4212cfdebf5be32dd
6 years ago
} else if (key == TablePropertiesNames::kCompressionOptions) {
new_table_properties->compression_options = raw_val.ToString();
} else {
// handle user-collected properties
new_table_properties->user_collected_properties.insert(
{key, raw_val.ToString()});
}
}
if (s.ok()) {
*table_properties = new_table_properties;
if (ret_block_handle != nullptr) {
*ret_block_handle = handle;
}
if (verification_buf != nullptr) {
size_t len = static_cast<size_t>(handle.size() + kBlockTrailerSize);
*verification_buf = rocksdb::AllocateBlock(len, memory_allocator);
if (verification_buf->get() != nullptr) {
memcpy(verification_buf->get(), block_contents.data.data(), len);
}
}
} else {
delete new_table_properties;
}
return s;
}
Status ReadTableProperties(RandomAccessFileReader* file, uint64_t file_size,
uint64_t table_magic_number,
const ImmutableCFOptions& ioptions,
TableProperties** properties,
bool compression_type_missing,
MemoryAllocator* memory_allocator) {
// -- Read metaindex block
Footer footer;
auto s = ReadFooterFromFile(file, nullptr /* prefetch_buffer */, file_size,
&footer, table_magic_number);
if (!s.ok()) {
return s;
}
auto metaindex_handle = footer.metaindex_handle();
BlockContents metaindex_contents;
ReadOptions read_options;
read_options.verify_checksums = false;
PersistentCacheOptions cache_options;
BlockFetcher block_fetcher(
file, nullptr /* prefetch_buffer */, footer, read_options,
metaindex_handle, &metaindex_contents, ioptions, false /* decompress */,
false /*maybe_compressed*/, BlockType::kMetaIndex,
UncompressionDict::GetEmptyDict(), cache_options, memory_allocator);
s = block_fetcher.ReadBlockContents();
if (!s.ok()) {
return s;
}
// property blocks are never compressed. Need to add uncompress logic if we
// are to compress it.
Block metaindex_block(std::move(metaindex_contents),
kDisableGlobalSequenceNumber);
Add an option to put first key of each sst block in the index (#5289) Summary: The first key is used to defer reading the data block until this file gets to the top of merging iterator's heap. For short range scans, most files never make it to the top of the heap, so this change can reduce read amplification by a lot sometimes. Consider the following workload. There are a few data streams (we'll be calling them "logs"), each stream consisting of a sequence of blobs (we'll be calling them "records"). Each record is identified by log ID and a sequence number within the log. RocksDB key is concatenation of log ID and sequence number (big endian). Reads are mostly relatively short range scans, each within a single log. Writes are mostly sequential for each log, but writes to different logs are randomly interleaved. Compactions are disabled; instead, when we accumulate a few tens of sst files, we create a new column family and start writing to it. So, a typical sst file consists of a few ranges of blocks, each range corresponding to one log ID (we use FlushBlockPolicy to cut blocks at log boundaries). A typical read would go like this. First, iterator Seek() reads one block from each sst file. Then a series of Next()s move through one sst file (since writes to each log are mostly sequential) until the subiterator reaches the end of this log in this sst file; then Next() switches to the next sst file and reads sequentially from that, and so on. Often a range scan will only return records from a small number of blocks in small number of sst files; in this case, the cost of initial Seek() reading one block from each file may be bigger than the cost of reading the actually useful blocks. Neither iterate_upper_bound nor bloom filters can prevent reading one block from each file in Seek(). But this PR can: if the index contains first key from each block, we don't have to read the block until this block actually makes it to the top of merging iterator's heap, so for short range scans we won't read any blocks from most of the sst files. This PR does the deferred block loading inside value() call. This is not ideal: there's no good way to report an IO error from inside value(). As discussed with siying offline, it would probably be better to change InternalIterator's interface to explicitly fetch deferred value and get status. I'll do it in a separate PR. Pull Request resolved: https://github.com/facebook/rocksdb/pull/5289 Differential Revision: D15256423 Pulled By: al13n321 fbshipit-source-id: 750e4c39ce88e8d41662f701cf6275d9388ba46a
6 years ago
std::unique_ptr<InternalIterator> meta_iter(metaindex_block.NewDataIterator(
BytewiseComparator(), BytewiseComparator()));
// -- Read property block
bool found_properties_block = true;
s = SeekToPropertiesBlock(meta_iter.get(), &found_properties_block);
if (!s.ok()) {
return s;
}
TableProperties table_properties;
if (found_properties_block == true) {
s = ReadProperties(
meta_iter->value(), file, nullptr /* prefetch_buffer */, footer,
ioptions, properties, false /* verify_checksum */,
nullptr /* ret_block_hanel */, nullptr /* ret_block_contents */,
compression_type_missing, memory_allocator);
} else {
s = Status::NotFound();
}
return s;
}
Status FindMetaBlock(InternalIterator* meta_index_iter,
const std::string& meta_block_name,
BlockHandle* block_handle) {
meta_index_iter->Seek(meta_block_name);
if (meta_index_iter->status().ok() && meta_index_iter->Valid() &&
meta_index_iter->key() == meta_block_name) {
Slice v = meta_index_iter->value();
return block_handle->DecodeFrom(&v);
} else {
return Status::Corruption("Cannot find the meta block", meta_block_name);
}
}
Status FindMetaBlock(RandomAccessFileReader* file, uint64_t file_size,
uint64_t table_magic_number,
const ImmutableCFOptions& ioptions,
const std::string& meta_block_name,
BlockHandle* block_handle,
bool /*compression_type_missing*/,
MemoryAllocator* memory_allocator) {
Footer footer;
auto s = ReadFooterFromFile(file, nullptr /* prefetch_buffer */, file_size,
&footer, table_magic_number);
if (!s.ok()) {
return s;
}
auto metaindex_handle = footer.metaindex_handle();
BlockContents metaindex_contents;
ReadOptions read_options;
read_options.verify_checksums = false;
PersistentCacheOptions cache_options;
BlockFetcher block_fetcher(
file, nullptr /* prefetch_buffer */, footer, read_options,
metaindex_handle, &metaindex_contents, ioptions,
false /* do decompression */, false /*maybe_compressed*/,
BlockType::kMetaIndex, UncompressionDict::GetEmptyDict(), cache_options,
memory_allocator);
s = block_fetcher.ReadBlockContents();
if (!s.ok()) {
return s;
}
// meta blocks are never compressed. Need to add uncompress logic if we are to
// compress it.
Block metaindex_block(std::move(metaindex_contents),
kDisableGlobalSequenceNumber);
std::unique_ptr<InternalIterator> meta_iter;
Add an option to put first key of each sst block in the index (#5289) Summary: The first key is used to defer reading the data block until this file gets to the top of merging iterator's heap. For short range scans, most files never make it to the top of the heap, so this change can reduce read amplification by a lot sometimes. Consider the following workload. There are a few data streams (we'll be calling them "logs"), each stream consisting of a sequence of blobs (we'll be calling them "records"). Each record is identified by log ID and a sequence number within the log. RocksDB key is concatenation of log ID and sequence number (big endian). Reads are mostly relatively short range scans, each within a single log. Writes are mostly sequential for each log, but writes to different logs are randomly interleaved. Compactions are disabled; instead, when we accumulate a few tens of sst files, we create a new column family and start writing to it. So, a typical sst file consists of a few ranges of blocks, each range corresponding to one log ID (we use FlushBlockPolicy to cut blocks at log boundaries). A typical read would go like this. First, iterator Seek() reads one block from each sst file. Then a series of Next()s move through one sst file (since writes to each log are mostly sequential) until the subiterator reaches the end of this log in this sst file; then Next() switches to the next sst file and reads sequentially from that, and so on. Often a range scan will only return records from a small number of blocks in small number of sst files; in this case, the cost of initial Seek() reading one block from each file may be bigger than the cost of reading the actually useful blocks. Neither iterate_upper_bound nor bloom filters can prevent reading one block from each file in Seek(). But this PR can: if the index contains first key from each block, we don't have to read the block until this block actually makes it to the top of merging iterator's heap, so for short range scans we won't read any blocks from most of the sst files. This PR does the deferred block loading inside value() call. This is not ideal: there's no good way to report an IO error from inside value(). As discussed with siying offline, it would probably be better to change InternalIterator's interface to explicitly fetch deferred value and get status. I'll do it in a separate PR. Pull Request resolved: https://github.com/facebook/rocksdb/pull/5289 Differential Revision: D15256423 Pulled By: al13n321 fbshipit-source-id: 750e4c39ce88e8d41662f701cf6275d9388ba46a
6 years ago
meta_iter.reset(metaindex_block.NewDataIterator(BytewiseComparator(),
BytewiseComparator()));
return FindMetaBlock(meta_iter.get(), meta_block_name, block_handle);
}
Status ReadMetaBlock(RandomAccessFileReader* file,
FilePrefetchBuffer* prefetch_buffer, uint64_t file_size,
uint64_t table_magic_number,
const ImmutableCFOptions& ioptions,
const std::string& meta_block_name, BlockType block_type,
BlockContents* contents, bool /*compression_type_missing*/,
MemoryAllocator* memory_allocator) {
Status status;
Footer footer;
status = ReadFooterFromFile(file, prefetch_buffer, file_size, &footer,
table_magic_number);
if (!status.ok()) {
return status;
}
// Reading metaindex block
auto metaindex_handle = footer.metaindex_handle();
BlockContents metaindex_contents;
ReadOptions read_options;
read_options.verify_checksums = false;
PersistentCacheOptions cache_options;
BlockFetcher block_fetcher(
file, prefetch_buffer, footer, read_options, metaindex_handle,
&metaindex_contents, ioptions, false /* decompress */,
false /*maybe_compressed*/, BlockType::kMetaIndex,
UncompressionDict::GetEmptyDict(), cache_options, memory_allocator);
status = block_fetcher.ReadBlockContents();
if (!status.ok()) {
return status;
}
// meta block is never compressed. Need to add uncompress logic if we are to
// compress it.
// Finding metablock
Block metaindex_block(std::move(metaindex_contents),
kDisableGlobalSequenceNumber);
std::unique_ptr<InternalIterator> meta_iter;
Add an option to put first key of each sst block in the index (#5289) Summary: The first key is used to defer reading the data block until this file gets to the top of merging iterator's heap. For short range scans, most files never make it to the top of the heap, so this change can reduce read amplification by a lot sometimes. Consider the following workload. There are a few data streams (we'll be calling them "logs"), each stream consisting of a sequence of blobs (we'll be calling them "records"). Each record is identified by log ID and a sequence number within the log. RocksDB key is concatenation of log ID and sequence number (big endian). Reads are mostly relatively short range scans, each within a single log. Writes are mostly sequential for each log, but writes to different logs are randomly interleaved. Compactions are disabled; instead, when we accumulate a few tens of sst files, we create a new column family and start writing to it. So, a typical sst file consists of a few ranges of blocks, each range corresponding to one log ID (we use FlushBlockPolicy to cut blocks at log boundaries). A typical read would go like this. First, iterator Seek() reads one block from each sst file. Then a series of Next()s move through one sst file (since writes to each log are mostly sequential) until the subiterator reaches the end of this log in this sst file; then Next() switches to the next sst file and reads sequentially from that, and so on. Often a range scan will only return records from a small number of blocks in small number of sst files; in this case, the cost of initial Seek() reading one block from each file may be bigger than the cost of reading the actually useful blocks. Neither iterate_upper_bound nor bloom filters can prevent reading one block from each file in Seek(). But this PR can: if the index contains first key from each block, we don't have to read the block until this block actually makes it to the top of merging iterator's heap, so for short range scans we won't read any blocks from most of the sst files. This PR does the deferred block loading inside value() call. This is not ideal: there's no good way to report an IO error from inside value(). As discussed with siying offline, it would probably be better to change InternalIterator's interface to explicitly fetch deferred value and get status. I'll do it in a separate PR. Pull Request resolved: https://github.com/facebook/rocksdb/pull/5289 Differential Revision: D15256423 Pulled By: al13n321 fbshipit-source-id: 750e4c39ce88e8d41662f701cf6275d9388ba46a
6 years ago
meta_iter.reset(metaindex_block.NewDataIterator(BytewiseComparator(),
BytewiseComparator()));
BlockHandle block_handle;
status = FindMetaBlock(meta_iter.get(), meta_block_name, &block_handle);
if (!status.ok()) {
return status;
}
// Reading metablock
BlockFetcher block_fetcher2(
file, prefetch_buffer, footer, read_options, block_handle, contents,
ioptions, false /* decompress */, false /*maybe_compressed*/, block_type,
UncompressionDict::GetEmptyDict(), cache_options, memory_allocator);
return block_fetcher2.ReadBlockContents();
}
} // namespace rocksdb