Re: Some performance questions

[cutting] Hi Doug, LRUS=3D=3D64 is a 'magic' number. =20 [cutting] Wasn't there a posting recently claiming 96 was a also a magic number? Or = am I going senile? Paul Watson WF Software Ltd Tel +44 1436 674729 Fax +44 1436 678729 www.wfsoftware.co.uk ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This footnote also confirms that this email message has been swept by MIMEsweeper for the presence of computer viruses. www.mimesweeper.com **********************************************************************

Paul watson wrote: > > [cutting] > > Hi Doug, LRUS=3D=3D64 is a 'magic' number. =20 > > [cutting] > > Wasn't there a posting recently claiming 96 was a also a magic number? Or = > am I going senile? There was one report of poor bufwaits performance at 96, I have not verified it myself. So far the known nodes are 64 and 96 with 16 being suspicious and one report of 32 being bad on 7.31 just this week. I hope that was a fluke or just a very busy system with too many users for 32 LRUS since 32 is a VERY popular value for LRUS. However, Menlo says the whole LRU hash/rehash code was rewritten for 7.31 so there is an outside possibility that the shifted the nodes. Art S. Kagel