Charles, In the past I found that BUFNO=16 was the knee in the curve for half track blocking on 3390. Anything beyond that gave pretty well no benefit, probably because 8x27998 would hit the chain limit, giving pretty close to half the buffers for IO while the other half are available for processing.
Ron > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[EMAIL PROTECTED] On > Behalf Of Charles Mills > Sent: Saturday, 27 August 2005 2:29 AM > To: IBM-MAIN@BAMA.UA.EDU > Subject: Most effective BUFNO? > > If you were reading from one QSAM DASD DCB and writing to another QSAM > DASD DCB -- let's say 20,000 133-byte records blocked half track -- > would you expect to see a noticeable performance improvement (elapsed > time and/or CPU time) from specifying a BUFNO greater than 5? What > number might you specify to optimize elapsed and/or CPU time without > "being ridiculous"? Any related considerations? > > Thanks, > > Charles Mills > > > > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO > Search the archives at http://bama.ua.edu/archives/ibm-main.html ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
