mirror of https://github.com/facebook/rocksdb.git
465b9103f8
Summary: Currently the test tracks all writes in memory and uses it for verification at the end. This has 4 problems: (a) It needs mutex for each write to ensure in-memory update and leveldb update are done atomically. This slows down the benchmark. (b) Verification phase at the end is time consuming as well (c) Does not test batch writes or snapshots (d) We cannot kill the test and restart multiple times in a loop because in-memory state will be lost. I am adding a FLAGS_multi that does MultiGet/MultiPut/MultiDelete instead of get/put/delete to get/put/delete a group of related keys with same values atomically. Every get retrieves the group of keys and checks that their values are same. This does not have the above problems but the downside is that it does less amount of validation than the other approach. Test Plan: This whole this is a test! Here is a small run. I am doing larger run now. [nponnekanti@dev902 /data/users/nponnekanti/rocksdb] ./db_stress --ops_per_thread=10000 --multi=1 --ops_per_key=25 LevelDB version : 1.5 Number of threads : 32 Ops per thread : 10000 Read percentage : 10 Delete percentage : 30 Max key : 2147483648 Num times DB reopens: 10 Num keys per lock : 4 Compression : snappy ------------------------------------------------ Creating 536870912 locks 2013/02/20-16:59:32 Starting database operations Created bg thread 0x7f9ebcfff700 2013/02/20-16:59:37 Reopening database for the 1th time 2013/02/20-16:59:46 Reopening database for the 2th time 2013/02/20-16:59:57 Reopening database for the 3th time 2013/02/20-17:00:11 Reopening database for the 4th time 2013/02/20-17:00:25 Reopening database for the 5th time 2013/02/20-17:00:36 Reopening database for the 6th time 2013/02/20-17:00:47 Reopening database for the 7th time 2013/02/20-17:00:59 Reopening database for the 8th time 2013/02/20-17:01:10 Reopening database for the 9th time 2013/02/20-17:01:20 Reopening database for the 10th time 2013/02/20-17:01:31 Reopening database for the 11th time 2013/02/20-17:01:31 Starting verification Stress Test : 109.125 micros/op 22191 ops/sec : Wrote 0.00 MB (0.23 MB/sec) (59% of 32 ops) : Deleted 10 times 2013/02/20-17:01:31 Verification successful Revert Plan: OK Task ID: # Reviewers: dhruba, emayanke Reviewed By: emayanke CC: leveldb Differential Revision: https://reviews.facebook.net/D8733 |
||
---|---|---|
db | ||
doc | ||
hdfs | ||
helpers/memenv | ||
include/leveldb | ||
java | ||
linters/src | ||
port | ||
scribe | ||
snappy | ||
table | ||
thrift | ||
tools | ||
util | ||
.arcconfig | ||
.gitignore | ||
LICENSE | ||
Makefile | ||
README | ||
README.fb | ||
build_detect_platform | ||
build_detect_version | ||
build_java.sh | ||
fbcode.clang31.sh | ||
fbcode.gcc471.sh | ||
regression_build_test.sh |
README
rocksdb: A persistent key-value store for flash storage Authors: The Facebook Database Engineering Team This code is a library that forms the core building block for a fast key value server, especially suited for storing data on flash drives. It has an Log-Stuctured-Merge-Database (LSM) design with flexible tradeoffs between Write-Amplification-Factor(WAF), Read-Amplification-Factor (RAF) and Space-Amplification-Factor(SAF). It has multi-threaded compactions, making it specially suitable for storing multiple terabytes of data in a single database. The core of this code has been derived from open-source leveldb. The code under this directory implements a system for maintaining a persistent key/value store. See doc/index.html for more explanation. See doc/impl.html for a brief overview of the implementation. The public interface is in include/*.h. Callers should not include or rely on the details of any other header files in this package. Those internal APIs may be changed without warning. Guide to header files: include/db.h Main interface to the DB: Start here include/options.h Control over the behavior of an entire database, and also control over the behavior of individual reads and writes. include/comparator.h Abstraction for user-specified comparison function. If you want just bytewise comparison of keys, you can use the default comparator, but clients can write their own comparator implementations if they want custom ordering (e.g. to handle different character encodings, etc.) include/iterator.h Interface for iterating over data. You can get an iterator from a DB object. include/write_batch.h Interface for atomically applying multiple updates to a database. include/slice.h A simple module for maintaining a pointer and a length into some other byte array. include/status.h Status is returned from many of the public interfaces and is used to report success and various kinds of errors. include/env.h Abstraction of the OS environment. A posix implementation of this interface is in util/env_posix.cc include/table.h include/table_builder.h Lower-level modules that most clients probably won't use directly