ceph/src/rocksdb/microbench/README.md

   1 # RocksDB Micro-Benchmark
   2
   3 ## Overview
   4
   5 RocksDB micro-benchmark is a set of tests for benchmarking a single component or simple DB operations. The test artificially generates input data and executes the same operation with it to collect and report performance metrics. As it's focusing on testing a single, well-defined operation, the result is more precise and reproducible, which also has its limitation of not representing a real production use case. The test author needs to carefully design the microbench to represent its true purpose.
   6
   7 The tests are based on [Google Benchmark](https://github.com/google/benchmark) library, which provides a standard framework for writing benchmarks.
   8
   9 ## How to Run
  10 ### Prerequisite
  11 Install the [Google Benchmark](https://github.com/google/benchmark) version `1.6.0` or above.
  12
  13 *Note: Google Benchmark `1.6.x` is incompatible with previous versions like `1.5.x`, please make sure you're using the newer version.*
  14
  15 ### Build and Run
  16 With `Makefile`:
  17 ```bash
  18 $ DEBUG_LEVEL=0 make run_microbench
  19 ```
  20 Or with cmake:
  21 ```bash
  22 $ mkdir build && cd build && cmake .. -DCMAKE_BUILD_TYPE=Release -DWITH_BENCHMARK
  23 $ make run_microbench
  24 ```
  25
  26 *Note: Please run the benchmark code in release build.*
  27 ### Run Single Test
  28 Example:
  29 ```bash
  30 $ make db_basic_bench
  31 $ ./db_basic_bench --benchmark_filter=<TEST_NAME>
  32 ```
  33
  34 ## Best Practices
  35 #### * Use the Same Test Directory Setting as Unittest
  36 Most of the Micro-benchmark tests use the same test directory setup as unittest, so it could be overridden by:
  37 ```bash
  38 $ TEST_TMPDIR=/mydata/tmp/ ./db_basic_bench --benchmark_filter=<TEST_NAME>
  39 ```
  40 Please also follow that when designing new tests.
  41
  42 #### * Avoid Using Debug API
  43 Even though micro-benchmark is a test, avoid using internal Debug API like TEST_WaitForRun() which is designed for unittest. As benchmark tests are designed for release build, don't use any of that.
  44
  45 #### * Pay Attention to Local Optimization
  46 As a micro-benchmark is focusing on a single component or area, make sure it is a key part for impacting the overall application performance.
  47
  48 The compiler might be able to optimize the code that not the same way as the whole application, and if the test data input is simple and small, it may be able to all cached in CPU memory, which is leading to a wrong metric. Take these into consideration when designing the tests.
  49
  50 #### * Names of user-defined counters/metrics has to be `[A-Za-z0-9_]`
  51 It's a restriction of the metrics collecting and reporting system RocksDB is using internally. It will also help integrate with more systems.
  52
  53 #### * Minimize the Metrics Variation
  54 Try reducing the test result variation, one way to check that is running the test multiple times and check the CV (Coefficient of Variation) reported by gbenchmark.
  55 ```bash
  56 $ ./db_basic_bench --benchmark_filter=<TEST_NAME> --benchmark_repetitions=10
  57 ...
  58 <TEST_NAME>_cv    3.2%
  59 ```
  60 RocksDB has background compaction jobs which may cause the test result to vary a lot. If the micro-benchmark is not purposely testing the operation while compaction is in progress, it should wait for the compaction to finish (`db_impl->WaitForCompact()`) or disable auto-compaction.