August 9, 2012 at 4:18 am
Hello guys,
I need you help with some advises about a high pageiolatch SH value.
So after I run Paul Randal wait script the report looks :
WaitType Wait_S Resource_S Signal_S WaitCount Percentage AvgWait_S AvgRes_SwA AvgSig_S
PAGEIOLATCH_SH 434855.86 434629.66 226.20 55031445 48.05 0.0079 0.0079 0.0000
ASYNC_NETWORK_IO 170614.17 169940.69 673.48 49831877 18.85 0.0034 0.0034 0.0000
ASYNC_IO_COMPLETION 86361.11 86361.06 0.05 72 9.54 1199.4598 1199.4592 0.0006
BACKUPBUFFER 85818.53 85784.05 34.48 2940403 9.48 0.0292 0.0292 0.0000
BACKUPIO 36485.20 36480.14 5.06 966040 4.03 0.0378 0.0378 0.0000
WRITELOG 23904.56 23850.88 53.69 3801876 2.64 0.0063 0.0063 0.0000
OLEDB 19365.61 19365.61 0.00 321505291 2.14 0.0001 0.0001 0.0000
PAGEIOLATCH_EX 18103.08 18097.58 5.50 1224365 2.00 0.0148 0.0148 0.0000
Env MSSQL 2005Sp4/Windows 2003Sp2 64b 32GB RAM total, 8GB dedicated to MSSQL,o n the same machine is SAP Netweaver installed.
Hardware SAN EMC RAID5 for Data ,RAID1 for Log file ,tempdb on the same drive like data ,backup occured on network share .
I have checked the HBA queue and looks OK.
August 9, 2012 at 11:21 am
A high percentage of PAGEIOLATCH_SH waits can sometimes be caused by insufficient memory to handle the workload or by poorly designed queries/indexes. Look at page life expectancy and target/total server memory and check for signs of memory pressure.
You can also check your datafile latency via perfmon counters, or using fn_virtualfilestats. Another query from Paul Randal. http://www.sqlskills.com/BLOGS/PAUL/post/How-to-examine-IO-subsystem-latencies-from-within-SQL-Server.aspx
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply