Add benchmark doc. (#335)

.gitignore

@@ -56,7 +56,7 @@
 
 # build directory of Gradle, CMake etc
 tmp
-build
+build*/
 cmake-build-*
 
 # ignore compilation configure file

docs/benchmark.md

@@ -0,0 +1,84 @@
+# Benchmark
+
+**infinity** supply python script for sift1m and gist1m dataset benchmark.
+
+## Get infinity binary file
+
+```sh
+git clone https://github.com/infiniflow/infinity.git
+cd infinity
+```
+
+## download benchmark file
+
+download via wget
+
+```sh
+#download sift benchmark
+wget ftp://ftp.irisa.fr/local/texmex/corpus/sift.tar.gz
+#download gist benchmark
+wget ftp://ftp.irisa.fr/local/texmex/corpus/gist.tar.gz
+
+```
+
+or  
+visit [http://corpus-texmex.irisa.fr/](http://corpus-texmex.irisa.fr/) to download manually.
+
+```sh
+#uncompress and move benchmark file
+tar -zxvf sift.tar.gz
+mv sift/sift_base.fvecs test/data/benchmark/sift_1m/sift_base.fvecs
+mv sift/sift_query.fvecs test/data/benchmark/sift_1m/sift_query.fvecs
+mv sift/sift_groundtruth.ivecs test/data/benchmark/sift_1m/sift_groundtruth.ivecs
+
+
+tar -zxvf gist.tar.gz
+mv gist/gist_base.fvecs test/data/benchmark/gist_1m/gist_base.fvecs
+mv gist/gist_query.fvecs test/data/benchmark/gist_1m/gist_query.fvecs
+mv gist/gist_groundtruth.ivecs test/data/benchmark/gist_1m/gist_groundtruth.ivecs
+
+```
+
+## Benchmark dependency
+
+```sh
+cd python
+
+pip install -r requirements.txt
+python setup.py bdist_wheel
+pip install dist/infinity_sdk-0.1.0.dev1-py3-none-any.whl
+```
+
+## Start infinity
+
+Read [README.md](https://github.com/infiniflow/infinity/blob/main/README.md) to start infinity.
+
+## Import data
+
+```sh
+cd benchmark
+
+options:
+  -h, --help            show this help message and exit
+  -d DATA_SET, --data DATA_SET
+
+python remote_benchmark_import.py -d sift_1m
+python remote_benchmark_import.py -d gist_1m
+```
+
+## run benchmark
+
+```sh
+options:
+  -h, --help            show this help message and exit
+  -t THREADS, --threads THREADS
+  -r ROUNDS, --rounds ROUNDS
+  -d DATA_SET, --data DATA_SET
+
+# ROUNDS means how many times python runs the benchmark. The result is the average for each time.
+
+# following command means run benchmark with 1 thread, for 1 time using sift dataset
+python remote_benchmark.py -t 1 -r 1 -d sift_1m
+
+python remote_benchmark.py -t 1 -r 1 -d gist_1m
+```