Revision 4a21b1402cb454f6e68d4c6f288d73700b3ff9db authored by Islam AbdelRahman on 01 December 2016, 21:30:01 UTC, committed by Facebook Github Bot on 01 December 2016, 21:39:14 UTC
Summary: Reduce number of comparisons in heap by caching which child node in the first level is smallest (left_child or right_child) So next time we can compare directly against the smallest child I see that the total number of calls to comparator drops significantly when using this optimization Before caching (~2mil key comparison for iterating the DB) ``` $ DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readseq" --db="/dev/shm/heap_opt" --use_existing_db --disable_auto_compactions --cache_size=1000000000 --perf_level=2 readseq : 0.338 micros/op 2959201 ops/sec; 327.4 MB/s user_key_comparison_count = 2000008 ``` After caching (~1mil key comparison for iterating the DB) ``` $ DEBUG_LEVEL=0 make db_bench -j64 && ./db_bench --benchmarks="readseq" --db="/dev/shm/heap_opt" --use_existing_db --disable_auto_compactions --cache_size=1000000000 --perf_level=2 readseq : 0.309 micros/op 3236801 ops/sec; 358.1 MB/s user_key_comparison_count = 1000011 ``` It also improves Closes https://github.com/facebook/rocksdb/pull/1600 Differential Revision: D4256027 Pulled By: IslamAbdelRahman fbshipit-source-id: 76fcc66
1 parent e39d080
verify_random_db.sh
#!/bin/bash
#
# A shell script to verify DB generated by generate_random_db.sh cannot opened and read correct data.
# ./ldb needs to be avaible to be executed.
#
# Usage: <SCRIPT> <DB Path>
scriptpath=`dirname $BASH_SOURCE`
if [ "$#" -lt 2 ]; then
echo "usage: $BASH_SOURCE <db_directory> <compare_base_db_directory> [dump_file_name]"
exit 1
fi
db_dir=$1
base_db_dir=$2
dump_file_name=${3:-"dump_file.txt"}
db_dump=$db_dir"/"$dump_file_name
base_db_dump=$base_db_dir"/"$dump_file_name
set -e
echo == Dumping data from $db_dir to $db_dump
./ldb dump --db=$db_dir > $db_dump
echo == Dumping data from $base_db_dir to $base_db_dump
./ldb dump --db=$base_db_dir > $base_db_dump
diff $db_dump $base_db_dir
Computing file changes ...
