… can assume that each random I/O has to wait for the head to move, for the disk to rotate the starting point… average) and then the disk to rotate all the data under the head. So the spindle rotation speed really dominates, and you can just estimate the random I/O off that. You can get…
Comment: The MySQL optimizer, the OS cache, and sequential versus random I/O
March 18, 2009 By 17 Comments
Comment: MySQL 6.0 vs 5.1 in TPC-H queries
June 21, 2008 By 17 Comments
… disk drive’s heads from place to place. As I first read about MRR and expectations this technique should provide the same speed advantage as clustered indexes, I laughed 
Actually MRR would help a little only because MySQL doesn’t take advantage from asynchronous I/O support (or am I wrong?). Having async I/O makes MRR useless, as the OS / disk drive’s firmware does…
Comment: Living with backups
November 18, 2008 By 28 Comments
… a single process saturating the server with I/O reads cause replicating to slightly lag behind. I can most likely avoid the checksums… read I/O in the Idle was making *both* dd (Idle class) and MySQL (default class) to slow down to ridiculously low speed… (3 stripped RAID1), 6x 10k SCSI-320 disks, 128M controller cache without write caching. I think that might be of interest to…
Comment: MySQL 5.6.7-RC in tpcc-mysql benchmark
October 10, 2012 By 22 Comments
…same drives as the data. So you get random i/o between them, flushing and foreground work. This is …different drives compared to using the same drives to speed up other things as well. It’s also …amounts of it unusable but still using RAM and disk space. For common production systems, the goal tends …
Post: To find the bottleneck, stop guessing and start measuring
August 4, 2008 By 8 Comments
We recently examined a customer’s system to try to speed up an ETL (Extraction, Transformation and Loading) process for a…, powerful) database server was almost completely idle, with virtually no I/O activity or CPU usage. So we looked at the…% CPU usage and was writing some files to disk, but not waiting on I/O. What’s going on here? Where’s…
Comment: Setting up XFS on Hardware RAID -- the simple edition
December 17, 2011 By 9 Comments
… which allows for non disruptive addition of disks and re-striping of data over more disks as you add them to the… i/o across every spindle in the system(assuming they are of the same speed e.g. 15k rpm 7200 rpm) by default(I…
Comment: Disaster: MySQL 5.5 Flushing
September 19, 2011 By 27 Comments
… have looked into, I do not think there linked lists or other bubble sorts or other funny O(2^n) problems in this case, it is just speed problem – changes in memory are faster than changes on disks. You… simple than current implementation may work on SSD. On spinning disks I feel it will be much problems, it will still take…
Post: Countless storage engines
May 3, 2007 By 6 Comments
… – transparent ZIP of pages stored on disk. Potentially it can imporove performance of tasks with intensive I/O load a lot. Second feature… very fast way aggregative functions on massive data sets. As I understood they use special index to store pre-calculated values… extension of “Patricia Tree” data structure. The developers promises ultimate speed and scalability for index access operations. It’s interesting to…
Post: MySQL 5.6 vs MySQL 5.5 and the Star Schema Benchmark
March 11, 2013 By 7 Comments
… 4G buffer pool instead of the default 128M Rationale: Since O_DIRECT is not used by default, the file system cache… disk problem, I pointed the MySQL 5.6.10 at the MySQL 5.5.30 data directory. Tthe speed was essentially the same. I did some further investigation and I determined that there was a lower…
Post: Content delivery system design mistakes
February 11, 2007 By 9 Comments
…content partition. Some web servers also support for O_NOATIME file open flag on newer Linux versions…use disk cache for static content as getting it from the server may be close in speed. It…a lot from one. Using different servers As I already mentioned serving different content requires different …
