Anti-Caching in the H-Store Parallel Main Memory OLTP Database System
C++ Java C Python Shell TeX Other
Clone or download
Pull request Compare This branch is 635 commits behind apavlo:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
properties
scripts
src
tests
third_party
tools
.classpath
.cproject
.gitignore
.project
COPYING
LICENSES
README.md
build-common.xml
build.py
build.xml
buildtools.py
experiment.sh
hstore
log4j.properties
run.sh
run_tpcc.sh
run_ycsb.sh
version.txt

README.md

H-Store

H-Store is an experimental main-memory, parallel database management system that is optimized for on-line transaction processing (OLTP) applications. It is a highly distributed, row-store-based relational database that runs on a cluster on shared-nothing, main memory executor nodes.

More information and documentation is available at: http://hstore.cs.brown.edu

Supported Platforms

H-Store is known to work on the following platforms. Please note that it will not compile on 32-bit systems.

  • Ubuntu Linux 9.10+ (64-bit)
  • Red Hat Enterprise Linux 5.5 (64-bit)
  • Mac OS X 10.6+ (64-bit)

Dependencies

Quick Start

  1. First build the entire distribution:

     ant build
    
  2. Next make the project jar file for the target benchmark. H-Store includes several benchmarks that are built-in and ready to execute. A project jar contains all the of stored procedures and statements for the target benchmark, as well as the cluster configuration for the database system.

     export HSTORE_BENCHMARK=tm1
     ant hstore-prepare -Dproject=$HSTORE_BENCHMARK
    
  3. You can now execute the benchmark locally on your machine with two partitions

     ant hstore-benchmark -Dproject=$HSTORE_BENCHMARK
    

More information is available here: http://hstore.cs.brown.edu/doc/quick-start/