Comparison sequential write
In Figure 1 a sequential-write 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.
1 through 2 workload jobs:
- SMB provides leading throughput, but NFS leads in CPU efficiency.
4 through 16 workload jobs:
- iSCSI and SMB shows leading peaks for throughput, but SMB exhibits leading values for CPU efficiency.
32 through 128 workload jobs:
- NFS provides leading throughput and CPU efficiency.