The wait stats on the database I am assigned for performance tuning , shows that the PAGEIOLATCH_SH and CXPACKETS share about 60% of the whole lot( ASYNC_NETWORK_IO too jumps sometimes but I checked it's due to the backups running at that time). Is this usual ? I mean is there any guideline which suggests certain threshold should be considered as not normal and looked-into ?
Basing on the findings in conjunction with the IO STALLS, I listed disk bottlenecks as an issue at the storage too as large number of 'pending-status' appears as 1 (pending) for the same files and for same sessions. For the CXPACKETS , parallelism reduction is also under consideration for some SPs . Any comments on this ?