Hello all, I am using libhdfs to generate io workload. I am trying to benchmark hdfs and ceph performance in IO workloads, But I am stuck accessing mixed read and write workloads. I have tried running them separately in two different jobs for hdfs as follows: hdfs.fio [global] namenode=10.4.xx.xx port=9000 hdfsdirectory=/user/hduser ioengine=libhdfs [hdfs-write] chunck_size=3m readwrite=write filesize=9m nrfiles=2 openfiles=1 [hdfs-read] chunck_size=1m readwrite=read filesize=1m nrfiles=1 But when I change the readwrite parameter to rw, I get hdfs error as below: [hdfs-write] chunck_size=3m readwrite=rw filesize=9m nrfiles=2 openfiles=1 Error shown is: Cannot write into a non-OutputStream object! Can someone help me, how you can generate random mixed read-write workload for both libhdfs and librbd ? Regards, -- To unsubscribe from this list: send the line "unsubscribe fio" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
