41 reputation
4
bio website
location
age
visits member for 2 years, 5 months
seen Dec 1 at 11:27

Apr
16
comment iozone: 5 concurrent threads random read of 300K blocks does more ops/s than using only one thread, what could be the underlying reason?
This might be correct. fread and read has a parameter which specifies file size, so hard drive controller can know for each block which clusters it has to read. When having multiple threads reading, it can benefit from optimizing the head movement between each blocks.
Apr
14
comment iozone: 5 concurrent threads random read of 300K blocks does more ops/s than using only one thread, what could be the underlying reason?
3.2.0-60-generic I think the underlying reason might be RAID1 with partition, with 2+ threads one thread reads from one drive, another from another drive (or something like that theoretically).
Apr
13
comment iozone: 5 concurrent threads random read of 300K blocks does more ops/s than using only one thread, what could be the underlying reason?
it's reading blocks of 300K. iozone creates files and read blocks of specific size, it is a HD benchmark software.
Apr
13
revised iozone: 5 concurrent threads random read of 300K blocks does more ops/s than using only one thread, what could be the underlying reason?
edited title
Apr
13
asked iozone: 5 concurrent threads random read of 300K blocks does more ops/s than using only one thread, what could be the underlying reason?
Apr
10
awarded  Teacher
Apr
10
answered How can I force users from USA to go the USA server and visitors from UK to go to the UK server
Apr
10
comment What are challenges with Linux Storage Servers having 7 or 15 drives?
images are 400KB in size, so according to this overclock.net/t/844026/average-7200rpm-read-speed maybe read speak is about 10MB/s-30MBs without RAID. Without actual testing I could only make rought guesses. Thanks for pointing too high HD speeds.
Apr
10
comment What are challenges with Linux Storage Servers having 7 or 15 drives?
servers with 7 or 15 drives are uncommon in standard dedicated offerings, if you go to dreamhost, godaddy or rackspace. you might not find it. The largest available I could find is on iweb with 8x4TB. But anyhow, if servers are replaceable among different providers it brings flexibility, that's what I'd call "commodity hardware" - whats easily available everywhere
Apr
10
revised What are challenges with Linux Storage Servers having 7 or 15 drives?
clarification
Apr
10
asked What are challenges with Linux Storage Servers having 7 or 15 drives?
Apr
9
comment DegradedArray event after rsync but later mdadm and smartctl do not show any issue
Anyhow, GREAT CATCH! I'm amazed what people do notice at serverfault!
Apr
9
comment DegradedArray event after rsync but later mdadm and smartctl do not show any issue
hi kranteg, yes you are correct. I figured that two days ago when that server completely failed - both drives had error, although sda had health 90% but server4you customer support decided to replace both drives (they were actually very good), so I had to make a backup and I'm just reinstalling servers. If I noticed on time that I'm looking at the wrong server, I would not have to go through this :-).
Apr
9
revised DegradedArray event after rsync but later mdadm and smartctl do not show any issue
solution
Apr
8
awarded  Student
Mar
20
awarded  Editor
Mar
20
comment DegradedArray event after rsync but later mdadm and smartctl do not show any issue
hi Peter, I added script, thank you if you can have a look again
Mar
20
revised DegradedArray event after rsync but later mdadm and smartctl do not show any issue
added actual script
Mar
19
asked DegradedArray event after rsync but later mdadm and smartctl do not show any issue
Jul
26
awarded  Tumbleweed