RE: SQL update vs RPG/DDS update

From: Wilt, Charles

I'm not sure what your problem is, Gene, but it is not that
dirty records are being hidden from users due to caching.
The beauty of the single level store is that there is no such
thing as a dirty record. When you access a record, you are
accessing address of the virtual memory location where it
resides and that memory is the same for all users.

Not quite true Joe.

Sounds like RPG may be blocking the i/o of the file. In which case, while
the program has done the
WRITE, the data is still in the program buffers and has not yet been
passed to the DB.

I though I was pretty clear about updates vs. writes, Charles. While writes
may be blocked, the random nature of the CHAIN/UPDATE pretty much precludes
any blocking. I suppose you MIGHT see blocking on an update primary file,
but I haven't tried that in a long time, and anyway that's not how I read
the original post. Gene specifically used the word "update" which I took to
mean updating a record, which in turn means CHAIN/UPDATE.

On the other hand I suppose it's possible that he was talking about locking
the file and doing a bunch of writes. In that case, you're absolutely
correct that the writes could be blocked (which I also said), and I would
definitely go with the FEOD strategy; that's what I use as opposed to the
FRCRATIO.

Joe