projects / apple / system_cmds.git / blame
| Commit | Line | Data |
|---|---|---|
| 1a7e3f61 A |
1 | .\" Copyright (c) 2013, Apple Inc. All rights reserved. |
| 2 | .\" | |
| 3 | .Dd Oct 9, 2013 | |
| 4 | .Dt IOSIM 1 | |
| 5 | .Sh NAME | |
| 6 | .Nm iosim | |
| 7 | .Nd A performance tool to simulate I/O workloads. | |
| 8 | .Sh SYNOPSIS | |
| 9 | .Pp | |
| 10 | .Nm iosim | |
| 11 | .Ar -c <number> | |
| 12 | Burst Count. No. of I/Os performed in an I/O burst. | |
| 13 | Default Value: 10 | |
| 14 | Valid Range: [0, INT_MAX] | |
| 15 | .Pp | |
| 16 | .Nm iosim | |
| 17 | .Ar -i <msecs> | |
| 18 | Inter Burst Duration. Amount of time each thread sleeps between bursts (-1 indicates random durations between 0-100 msecs) | |
| 19 | Default Value: 0 | |
| 20 | Valid Range: [-1, INT_MAX] | |
| 21 | .Pp | |
| 22 | .Nm iosim | |
| 23 | .Ar -d <msecs> | |
| 24 | Inter I/O delay. Amount of time thread sleeps between issuing I/Os | |
| 25 | Default Value: 0 | |
| 26 | Valid Range: [0, INT_MAX] | |
| 27 | .Pp | |
| 28 | .Nm iosim | |
| 29 | .Ar -t <number> | |
| 30 | Thread count | |
| 31 | Default Value: 1 | |
| 32 | Valid Range: [0, 1000] | |
| 33 | .Pp | |
| 34 | .Nm iosim | |
| 35 | .Ar -f <number> | |
| 36 | Workload Type (0/1/2 : Read-Only/Write-Only/Mixed RW) | |
| 37 | Default Value: 0 | |
| 38 | Valid Range: [0, 2] | |
| 39 | .Pp | |
| 40 | .Nm iosim | |
| 41 | .Ar -m <number> | |
| 42 | I/O Pattern (0/1 : Sequential/Random) | |
| 43 | Default Value: 0 | |
| 44 | Valid Range: [0, 1] | |
| 45 | .Pp | |
| 46 | .Nm iosim | |
| 47 | .Ar -j <bytes> | |
| 48 | Size of I/O in bytes | |
| 49 | Default Value: 4096 | |
| 50 | Valid Range: [0, INT_MAX] | |
| 51 | .Pp | |
| 52 | .Nm iosim | |
| 53 | .Ar -s <msecs> | |
| 54 | Frequency of sync() calls | |
| 55 | Default Value: 5000 | |
| 56 | Valid Range: [0, INT_MAX] | |
| 57 | .Pp | |
| 58 | .Nm iosim | |
| 59 | .Ar -l <number> | |
| 60 | I/O Tier (0/1/2/3) | |
| 61 | Default Value: 0 | |
| 62 | Valid Range: [0, 3] | |
| 63 | .Pp | |
| 64 | .Nm iosim | |
| 65 | .Ar -z <number> | |
| 66 | Size of the file created specified in pages (Only used when the file is being created by the tool) | |
| 67 | Default Value: 1 GB | |
| 68 | Valid Range: [0, INT_MAX] | |
| 69 | .Pp | |
| 70 | .Nm iosim | |
| 71 | .Ar -x <secs> | |
| 72 | Test duration (0 indicates that the tool would wait for a Ctrl-C) | |
| 73 | Default Value: 0 | |
| 74 | Valid Range: [0, INT_MAX] | |
| 75 | .Pp | |
| 76 | .Nm iosim | |
| 77 | .Ar -a number | |
| 78 | I/O Caching behavior (0/1 : Non-cached/Cached) | |
| 79 | Default Value: 0 | |
| 80 | Valid Range: [0, 1] | |
| 81 | .Pp | |
| 82 | .Nm iosim | |
| 83 | .Ar -n <filename> | |
| 84 | Filename for I/Os (If this option is not specified, the tool would create files on its own) | |
| 85 | Valid Range: Valid filename | |
| 86 | .Sh DESCRIPTION | |
| 87 | The | |
| 88 | .Nm iosim | |
| 89 | tool allows simulating workloads for I/O performance evaluation. The tool spawns 'n' threads which issue non-cached I/Os. If specified, it also creates a sync thread which issues system wide sync() calls to flush data and metadata to disk (emulates launchd behavior). The I/Os are issued at the specified I/O tier and the tool reports latency and throughput numbers. | |
| 90 | .P | |
| 91 | .nf | |
| 92 | Following is an explanation of the results: | |
| 93 | Avg. Latency : Avg. latency experienced by the I/Os. | |
| 94 | Low Latency Histogram: Frequency distribution of I/O latency for low latency I/Os. | |
| 95 | Latency Histogram: Frequency distribution of I/O latency. | |
| 96 | Burst Avg. Latency Histogram: Frequency distribution of burst avg. latency. | |
| 97 | Throughput timeline: Time windowed throughput distrbution. | |
| 98 | .fi | |
| 99 | .Sh SEE ALSO | |
| 100 | .Xr fs_usage 1 |