In a message dated 4/25/2006 8:02:22 A.M. Central Daylight Time, [EMAIL PROTECTED] writes:
>Who knows how much collection costs? There are three components of collection that cause overhead directly attributable to SMF, but most of which is not measurable: (1) creation of data elements that are not functionally required for the proper functioning of the software component involved and are maintained only to be dumped into the SMF record; (2) gathering of all the data elements and building the record; (3) requesting the SMF writing service (used to be only via an SVC, now there is also branch entry or maybe PC). Once you are inside the writing service, the test is made as to whether or not that particular record type is to be written (true originally for all record types; now I am not sure if some can be turned off earlier in the process). IBM has some undocumented monitor call instructions inside the I/O first level interrupt handler for instrumentation purposes. Maybe they also have some way to measure the overhead of building an SMF record. And maybe they can report on it some day if they do. There is no way we can find out. The only part of the overhead we can measure is that belonging to the SMF writing service in the SMF address space, and that's a small part of the total. Bill Fairchild ---------------------------------------------------------------------- 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
