cassandra 集群并发测试脚本

prepare:

create keyspace ycsb  WITH REPLICATION = {'class' : 'SimpleStrategy', 'replication_factor': 2 };
USE ycsb;    
CREATE TABLE users (
 firstname text, 
 lastname text, 
 age int, 
 email text, 
 city text, 
 PRIMARY KEY (lastname)
);    

用bash开5个python进程来测试写入性能，注意：当集群数目增加，应当增加bash脚本的并发进程数！

test_cass.py

from cassandra.cluster import Cluster
from cassandra.query import BatchStatement

from time import time

cluster = Cluster(["10.178.204.225"])
#session = cluster.connect('my_keyspace')
session = cluster.connect('ycsb')
#session.execute("TRUNCATE my_keyspace.users;")
session.execute("TRUNCATE ycsb.users;")

start = time()
data_cnt = 1000000
bulk_cnt = 100
for i in range(data_cnt/bulk_cnt):
    insert_user = session.prepare("insert into users (lastname, age, city, email, firstname) values (?, ?, 'Austin', 'bob@example.com', 'Bob')")
    batch = BatchStatement()
    for j in range(bulk_cnt):
        batch.add(insert_user, ("Jones"+str(i*bulk_cnt+j), j))
    session.execute(batch)

end = time()
interval = end-start
print "time:", interval, "through ops/per second:", data_cnt/interval

result = session.execute("select * from users where lastname='Jones0'")
for x in result:
    print x.age, x.firstname, x.lastname
#result = session.execute("select count(*) from users")
#for x in result:
#    print x

result = session.execute("select * from users where lastname='Jones%s'" % (data_cnt-1))
for x in result:
    print x.age, x.firstname, x.lastname

bash run.sh

for((i=0;i<5;i++)) do
    python test_cass.py &
done
wait
echo "done"