IBM Mainframe Forum Index
 
Log In
 
IBM Mainframe Forum Index Mainframe: Search IBM Mainframe Forum: FAQ Register
 

Extracting Compression Statistics from DCOLLECT


IBM Mainframe Forums -> PL/I & Assembler
Post new topic   Reply to topic
View previous topic :: View next topic  
Author Message
Jim Stout

New User


Joined: 11 May 2013
Posts: 4
Location: USA

PostPosted: Wed Apr 27, 2022 3:44 am
Reply with quote

Hi Folks! This is a follow-on to an earlier question. This one has me puzzled.

I'm trying to extract compression statistics from a DCOLLECT but the fields and values are contradicting each other. I'm looking at the D-Records in the DCOLLECT.

When I look at DCDFLAG3 X'04' says that the values in DCDCTYPE (+385) are valid. There are four options for DCDCTYPE:
X'00' = Not Compressed
X'01' = Generic Compression
X'02" = Tailored Compression
x'03' = ZEDC

So far so good but when I look at DCDUSIZ (+264) and DCDCUDSZ (+272) these fields contain 8 byte values for the User Data Bytes (DCDUSIZ) and the Compressed Dataset Size (DCDUDSX). There is also a DCDEXFLG (+280) when bit X'80' is on, these size fields are invalid.

I have many DCOLLECTs from many different sites. When I find datasets that contain both User Data Bytes and Compressed Data Bytes with the flag indicating they are valid, the COMPTYPE field shows Not Compressed. Here are my questions:

Is DCDCTYPE valid and to be trusted? It's fairly new.

Are DCDUSIZ and DCDCUSZ only used for software compression and not ZEDC? If the dataset is compressed, regardless of the mechanism, shouldn't these fields contain data?

Does anyone have the full description of DCDEXFLG? The only bit setting I can find is the X'80' bit that says the DCDUSIZ and DCDCUSZ fields are invalid. Surely there are other meanings because I can see other bits set.

This is a lengthy post and fairly complex so I appreciate any insight into this.
Back to top
View user's profile Send private message
Pete Wilson

Active Member


Joined: 31 Dec 2009
Posts: 582
Location: London

PostPosted: Thu Apr 28, 2022 12:02 pm
Reply with quote

There is an APAR related to this

www.ibm.com/support/pages/apar/OA54897
Back to top
View user's profile Send private message
Jim Stout

New User


Joined: 11 May 2013
Posts: 4
Location: USA

PostPosted: Thu Apr 28, 2022 6:42 pm
Reply with quote

Thanks Pete! I've seen that one and a few others as well. The thing I'm trying to resolve is when there are values in DCDUSIZ and DCDCUDSZ and the DCDFLAG3 indicates these fields contain valid data, the (recently added) DCDCTYPE flag indicated "Not Compressed." The other thing I'm seeing, and this occurs mostly when encryption is also being used, is DCDCTYPE will show "Generic Compression" (X'01') but the DCDUSIZ and DCDCUDSZ fields contain zeros. It looks like DCDCTYPE is only being populated by some compression techniques but not all.
Back to top
View user's profile Send private message
View previous topic :: :: View next topic  
Post new topic   Reply to topic View Bookmarks
All times are GMT + 6 Hours
Forum Index -> PL/I & Assembler

 


Similar Topics
Topic Forum Replies
No new posts Extracting Variable decimal numbers f... DFSORT/ICETOOL 17
No new posts Help in extracting data between doubl... DFSORT/ICETOOL 5
No new posts DB2 Statistics - Using EXPLAIN and qu... DB2 1
No new posts Allocated space calculation from DCOL... PL/I & Assembler 3
No new posts Program level statistics CICS 6
Search our Forums:

Back to Top