зеркало из https://github.com/Azure/YCSB.git
12ce3a8504 | ||
---|---|---|
.. | ||
src | ||
README.md | ||
pom.xml |
README.md
Overview
GridDB (https://github.com/griddb/griddb_nosql) is a highly scalable NoSQL database best suited for IoT and Big Data.
This is GridDB Binding for YCSB.
Environment
Library building and program execution have been checked in the following environment.
OS: CentOS 6.9(x64).
Java: JDK 1.8.0_191
Maven: 3.5.4
Python: 2.6.6
Quick start
Preparations
Clone the YCSB source code from git repository:
git clone https://github.com/brianfrankcooper/YCSB.git
cd YCSB
Build
Run following command
$ mvn -pl site.ycsb:griddb-binding -am clean package
Then, some jar files are created in:
griddb/target
GridDB setup
Please set the number of cpu core as /dataStore/concurrency in gs_node.json and "32KB" as /dataStore/storeBlockSize in gs_cluster.json.
Run YCSB
GridDB needs to be started in advance.
Please execute our program with fieldcount=10 and fieldlength=100.
First, load the data:
./bin/ycsb load griddb -P workloads/workloada
-p notificationAddress=<GridDB notification address(default is 239.0.0.1)>
-p notificationPort=<GridDB notification port(default is 31999)>
-p clusterName=<GridDB cluster name>
-p userName=<GridDB user name>
-p password=<GridDB password>
-p fieldcount=10
-p fieldlength=100
Then, run the workload:
./bin/ycsb run griddb -P workloads/workloada
-p notificationAddress=<GridDB notification address(default is 239.0.0.1)>
-p notificationPort=<GridDB notification port(default is 31999)>
-p clusterName=<GridDB cluster name>
-p userName=<GridDB user name>
-p password=<GridDB password>
-p fieldcount=10
-p fieldlength=100