Hello folks, Perf tuning isn't my strong point so I'm looking for any pointers going. I have a query that is taking a couple of seconds to run on serverA yet over a minute on serverB. The underlying data is the same on both schemas, as are the tables & indexes. Here's the output from SET STATISTICS IO ON. First, serverA:Table 'Journal'. Scan count 9, logical reads 388397, physical reads 0, read-ahead reads 0, lob logical reads 0, lob physical reads 0, lob read-ahead reads 0. And now serverB:Table 'Journal'. Scan count 9, logical reads 381755, physical reads 5639, read-ahead reads 372576, lob logical reads 0, lob physical reads 0, lob read-ahead reads 0. As you can see physical reads and read-aheads are different! Note also, if I executeBCC DROPCLEANBUFFERS on serverA prior to running the query then the query takes abotu a minute yet on subsequent executions its back down to a couple of seconds. I am assuming that the data cache is being populated on serverA but not on serverB, and this is where I draw a blank. Why might this be? Like I say, any pointers greatly appreciated. Now going off to search for info on the data cache, but thought I'd ask here first in case anyone can help.... -Jamie More info: I've queried sys.dm_os_sys_info and discovered that serverA has bpool_commit_target=1036288, on serverB bpool_commit_target=524288. Not sure if this is significant or not. physical_memory_in_bytes and virtual_memory_in_bytes are the same on both servers.