Esteemed listers I was asked by many people about extents in DB2 tablespace. From literature and a wide number of observation in many shops I see that now with modern disk systems it doesn't matter anymore and it's stated in many papers,articles etc..
I saw big tablespace with 100+ extents working very well in disks were VTOC and ICF catalogs were defined in many ways. I never detected any degradation or problem i.e. I never saw and had (I worked and I work as a storage manager as well) had a VTOC full or ICF catalog suffering the fact there are many DB2 datasets with a lot of extents (as they are recorded in VTOC and catalog entries) . I never saw any wait on catalog due to too many extents (I saw other kinds of waits). Recently I was blamed that all is wrong for the following reasons: 1) 'Too information on VTOC due to extents. This could create problems.' - Actually I never saw a VTOC full error for this reason, only for many small datasets generated by particular products that created a huge number of VTOC entries. Or for VTOCs really small or defined in a very bad manner. I mean that the critical factor is the number of dataset, not extents. 2) ' Catalog can suffer as they have to register all extents.' - Never seen it as well. I saw some other probems but no one was due to extents number registered in the catalog. I'd have expected some doubts on extents chain search or something else. The person didn't mentioned VVDS as well It seems to me that the previous are arguments like define VTOC in the 'middle' of a disk or using imbed/replicate of some years ago. Anyway, Is there anyone out there who had problems due to 1) and/or 2) reasons ? Thank you in advance Max Scarpa DB2 sysprog ---------------------------------------------------------------------- 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
