Comparison random read

Edit online

In Figure 1 a random-read FIO workload is running against a file system. The workload is scaled from 1 to 128 jobs. Values for each of the protocols (iSCSI, NFS, SMB) are plotted. The throughput plot is normalized to 1 SMB job.

Figure 1. Comparison - random read
Comparison - random read

1 through 4 workload jobs:

  • Closely clustered throughput values with SMB leading in CPU efficiency.

8 through 32 workload jobs:

  • Increasing throughput values with iSCSI leading at the higher workload job counts. NFS exhibiting leading CPU efficiency at the higher workload job counts.

64 through 128 workload jobs:

  • SMB showing leading behavior in throughput and CPU efficiency.