On Sat, 29 Dec 2007, dean gaudet wrote:
On Tue, 25 Dec 2007, Bill Davidsen wrote:
The issue I'm thinking about is hardware sector size, which on modern drives
may be larger than 512b and therefore entail a read-alter-rewrite (RAR) cycle
when writing a 512b block.
i'm not sure any shipping SATA disks have larger than 512B sectors yet...
do you know of any? (or is this thread about SCSI which i don't pay
attention to...)
on a brand new WDC WD7500AAKS-00RBA0 with this partition layout:
255 heads, 63 sectors/track, 91201 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
so sda1 starts at a non-multiple of 4096 into the disk.
i ran some random seek+write experiments using
<http://arctic.org/~dean/randomio/>, here are the results using 512 byte
and 4096 byte writes (fsync after each write), 8 threads, on sda1:
# ./randomio /dev/sda1 8 1 1 512 10 6
total | read: latency (ms) | write: latency (ms)
iops | iops min avg max sdev | iops min avg max sdev
--------+-----------------------------------+----------------------------------
148.5 | 0.0 inf nan 0.0 nan | 148.5 0.2 53.7 89.3 19.5
129.2 | 0.0 inf nan 0.0 nan | 129.2 37.2 61.9 96.7 9.3
131.2 | 0.0 inf nan 0.0 nan | 131.2 40.3 61.0 90.4 9.3
132.0 | 0.0 inf nan 0.0 nan | 132.0 39.6 60.6 89.3 9.1
130.7 | 0.0 inf nan 0.0 nan | 130.7 39.8 61.3 98.1 8.9
131.4 | 0.0 inf nan 0.0 nan | 131.4 40.0 60.8 101.0 9.6
# ./randomio /dev/sda1 8 1 1 4096 10 6
total | read: latency (ms) | write: latency (ms)
iops | iops min avg max sdev | iops min avg max sdev
--------+-----------------------------------+----------------------------------
141.7 | 0.0 inf nan 0.0 nan | 141.7 0.3 56.3 99.3 21.1
132.4 | 0.0 inf nan 0.0 nan | 132.4 43.3 60.4 91.8 8.5
131.6 | 0.0 inf nan 0.0 nan | 131.6 41.4 60.9 111.0 9.6
131.8 | 0.0 inf nan 0.0 nan | 131.8 41.4 60.7 85.3 8.6
130.6 | 0.0 inf nan 0.0 nan | 130.6 41.7 61.3 95.0 9.4
131.4 | 0.0 inf nan 0.0 nan | 131.4 42.2 60.8 90.5 8.4
i think the anomalous results in the first 10s samples are perhaps the drive
coming out of a standby state.
and here are the results aligned using the sda raw device itself:
# ./randomio /dev/sda 8 1 1 512 10 6
total | read: latency (ms) | write: latency (ms)
iops | iops min avg max sdev | iops min avg max sdev
--------+-----------------------------------+----------------------------------
147.3 | 0.0 inf nan 0.0 nan | 147.3 0.3 54.1 93.7 20.1
132.4 | 0.0 inf nan 0.0 nan | 132.4 37.4 60.6 91.8 9.2
132.5 | 0.0 inf nan 0.0 nan | 132.5 37.7 60.3 93.7 9.3
131.8 | 0.0 inf nan 0.0 nan | 131.8 39.4 60.7 92.7 9.0
133.9 | 0.0 inf nan 0.0 nan | 133.9 41.7 59.8 90.7 8.5
130.2 | 0.0 inf nan 0.0 nan | 130.2 40.8 61.5 88.6 8.9
# ./randomio /dev/sda 8 1 1 4096 10 6
total | read: latency (ms) | write: latency (ms)
iops | iops min avg max sdev | iops min avg max sdev
--------+-----------------------------------+----------------------------------
145.4 | 0.0 inf nan 0.0 nan | 145.4 0.3 54.9 94.0 20.1
130.3 | 0.0 inf nan 0.0 nan | 130.3 36.0 61.4 92.7 9.6
130.6 | 0.0 inf nan 0.0 nan | 130.6 38.2 61.2 96.7 9.2
132.1 | 0.0 inf nan 0.0 nan | 132.1 39.0 60.5 93.5 9.2
131.8 | 0.0 inf nan 0.0 nan | 131.8 43.1 60.8 93.8 9.1
129.0 | 0.0 inf nan 0.0 nan | 129.0 40.2 62.0 96.4 8.8
it looks pretty much the same to me...
-dean
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Good to know/have it confirmed by someone else, the alignment does not
matter with Linux/SW RAID.
Justin.
-
To unsubscribe from this list: send the line "unsubscribe linux-raid" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
