Tobias, What sort of benefit were you expecting for sequential access? The default for sequential pre-fetch I believe is half a CA.
For random, don't you think 1MB is pretty small? Ron > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of > Tobias Cafiero > Sent: Tuesday, September 14, 2010 9:44 AM > To: [email protected] > Subject: [IBM-MAIN] DFSMS and System Managed Buffering > > Hello, > I'm testing the SMB option in the DFSMS DATACLS Constructs, but don't > yet see a performance boost as promised. At this point I'm using just a > del/define,repro and pointing to the SMB DATACLS. The DATACLS is extended > and contains the following values: > > Record Access Bias . . . . : SYSTEM > System Managed Buffer . . . : 1M > > The STORCLS is our standard non-striped type. Has anyone implemented SMB > on their system. > > Thanks in Advance > Tobias Cafiero > > ---------------------------------------------------------------------- > 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
