Appendix D. Performance, Test and Debugging tools

scu.  The SCSI Command Utility (SCU) implements various SCSI commands necessary for normal maintenance and diagnostics of SCSI peripherals. Some of its features include: formatting, scanning for (and reassigning) bad blocks, downloading new firmware, executing diagnostics and obtaining performance information. It is available on several Unix platforms (and NT), however it is only currently available in binary form. See www.bit-net.com/~rmiller/scu.html for more details.

dd.  Very useful for testing the streaming performance of disks and cdroms/dvds. See man dd for more details. Here is an example for timing how long a disk takes to read 1 GB (10**9 bytes) starting from block 0:

$ time dd if=/dev/sda of=/dev/null bs=512 count=1953126

If the raw device /dev/raw/raw1 is bound to /dev/sda then the above line is equivalent to:

$ time dd if=/dev/raw/raw1 of=/dev/null bs=512 count=1953126

This may be slower than expected since one 512 byte sector is being read at a time. Changing the last 2 arguments to "bs=8k count=122071" should give better timings for the "raw" dd.

dt.  The Data Test (DT) program is modelled on dd's syntax but dt can do a lot more than sequential copies. It is a comprehensive data test program for SCSI devices such as disks, tapes and cdrom/dvds. It is available on several Unix platforms (and NT), and its source is available (unlike its stable mate "scu" discussed earlier). See www.bit-net.com/~rmiller/dt.html for more details.

lmdd.  This command is part of the lmbench suite of programs and is a variant of the dd command. It has been tailored for IO measurements and outputs timing and throughput numbers on completion. Hence the time command and a calculator are not needed.

blockdev.  Fetches the sector size, the number of sectors and read ahead status of a block device (typically a disk). Can also be used to flush buffers and reread the partition table. See man blockdev.

sg_dd.  This command is part of the sg_utils package (see W4) and is another variant of the dd command in which either the input and/or output file is a sg or a raw device. The block size argument ("bs") must match that of the physical device in question. The "skip" and "seek" arguments can be up to 2**31 - 1 on a 32 bit architecture allowing 1TB disks to be accessed (2G * 512). The Linux system command llseek() is used to seek with a 64 bit file read/write offset. The lmdd does not handle the > 2GB case and the dd command gets creative with multiple relative seeks. sg_dd has a "bpt" (blocks per transfer) argument that controls the number of blocks read or written in each IO transaction.

There are other programs in the sg_utils package to scan the SCSI bus (sg_scan and sg_map), to measure SCSI bus throughput (sg_rbuf and sg_turs ), show data from the SCSI inquiry command (sg_inq) and spin up (or down) media (sg_start).

dd_rescue + scsiinfo.  This dd variant is designed to rescue damaged media such as SCSI (or IDE) disks and CDROMs (see W6). The scsiinfo utility for displaying and changing mode page information is also at that site.

sard.  This utility is modelled on System V Release 4's sar -d for producing IO statistics for mounted devices and partitions. It has been developed by Stephen Tweedie and includes the sard utility and a required kernel patch which expands the output of /proc/partitions . It can be found at ftp.uk.linux.org/pub/linux/sct/fs/profiling. It collects statistics at a relatively low level (e.g. SCSI mid level) compared to programs like vmstat (see "man vmstat").