Table of Contents

IOR

IOR is a synthetic benchmark for testing the performance of distributed filesystems. The benchmark supports a variety of different APIs to simulate IO load.

Usage

You can set the API to be used using the -a flag. A multitude of other commandline options is documented in ior's user guide.

Here are a few important ones:

Example Output

For example runnning ior -a POSIX (using ior in version 3.0.1) results in the following output

IOR-3.0.1: MPI Coordinated Test of Parallel I/O

Began: Mon Mar 21 16:19:50 2016
Command line used: ./ior -a POSIX
Machine: Linux hostname

Test 0 started: Mon Mar 21 16:19:50 2016
Summary:
	api                = POSIX
	test filename      = testFile
	access             = single-shared-file
	ordering in a file = sequential offsets
	ordering inter file= no tasks offsets
	clients            = 1 (1 per node)
	repetitions        = 1
	xfersize           = 262144 bytes
	blocksize          = 1 MiB
	aggregate filesize = 1 MiB

access    bw(MiB/s)  block(KiB) xfer(KiB)  open(s)    wr/rd(s)   close(s)   total(s)   iter
------    ---------  ---------- ---------  --------   --------   --------   --------   ----
write     2551.28    1024.00    256.00     0.000019   0.000372   0.000001   0.000392   0   
read      9800       1024.00    256.00     0.000001   0.000101   0.000000   0.000102   0   
remove    -          -          -          -          -          -          0.000156   0   

Max Write: 2551.28 MiB/sec (2675.21 MB/sec)
Max Read:  9799.78 MiB/sec (10275.81 MB/sec)

Summary of all tests:
Operation   Max(MiB)   Min(MiB)  Mean(MiB)     StdDev    Mean(s) Test# #Tasks tPN reps fPP reord reordoff reordrand seed segcnt blksiz xsize aggsize API RefNum
write        2551.28    2551.28    2551.28       0.00    0.00039 0 1 1 1 0 0 1 0 0 1 1048576 262144 1048576 POSIX 0
read         9799.78    9799.78    9799.78       0.00    0.00010 0 1 1 1 0 0 1 0 0 1 1048576 262144 1048576 POSIX 0

Finished: Mon Mar 21 16:19:50 2016

Output has been created using version 3.0.1.