home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: vmsnet.internals
- Path: sparky!uunet!decwrl!pa.dec.com!ripple.enet.dec.com!pettigrew_mi
- From: pettigrew_mi@ripple.enet.dec.com (Mike Pettigrew)
- Subject: Re: RE: VMS tuning for a vaxcluster (LAVC)
- Message-ID: <1992Aug29.013855.12775@PA.dec.com>
- Sender: news@PA.dec.com (News)
- Organization: Digital Equipment Corporation
- References: <9208281916.AA02286@sndsu1.sinet.slb.com>
- Date: 28 AUG 92 17:53:34 PDT
- Lines: 91
-
-
- In article <9208281916.AA02286@sndsu1.sinet.slb.com>,
- <brydon@dsny25.sinet.slb.com> writes...
-
- >
- >Okay, but what about the situation of one satellite vaxstation on a lightly
- >loaded network? I think page read I/O's should be considered separately from
- >page write I/O's. Does this satellite system do page read I/O's any faster
- >over the ethernet versus to local disk?
- >
-
- Paging I/O takes longer from a server over the ethernet than it does from
- a local disk. There is not an easy way to separately measure page read and
- page write I/O times, nor any obvious reason why they should be greatly
- different. No one has yet noticed if this is so, though it is certainly
- possible.
-
-
- A satellite system that is paging any processes always seems to run slower
- with a server page file than with a local page file, even when the net is
- lightly loaded. You can measure the difference in a very straightforward
- manner.
-
- It is easy to measure program run time with a server page file and then
- again with a local page file. It is somwhat difficult to make sure other
- conditions remain constant, but when you do, the answers are consistent.
-
-
-
- >
- >>What does have a very critical effect is changes to the system parameters
- >>WSMAX, MPW_HILIMIT, PQL_MWSQUOTA, PQL_MWSEXTENT , and also changes to the
- >>SYSUAF field WSEXTENT. On a memory-rich system, these fields can be raised
- >>to much higher values than the default calculations.
- >
- >Hmmmm well agreed, it's like money. If you have more you can spend more. In
- >my opinion Autogen does a good enough job of this except of course for
- >WSEXTENT and perhaps PQL_MWSEXTENT.
- >
- >>The practical effect of raising these parameters is to eliminate paging
- >>and swapping activity on the sattelite nodes althogether.
- >>
- >>If you INSTALL all of the images that you are using on the sattelite nodes,
- >>and keep a free list of about 4,000 pages, you will rapidly have all of
- >>the relavent system image pages cached in the free list at the sattelite
- >>node.
- >
- >This sounds a bit like the old DEC sales pitch of "buy more memory". Indeed
- >it will improve matters, but that wasn't really the question being raised.
- >
-
- Two questions were asked:
-
- (1) Do page reads work faster from a server than from a local disk
- if data blocks are cached?
-
- No, because you cannot cache the data blocks on the page file.
- The local page file is always faster, and measurably so, given
- equal disk drive speeds.
-
- If your application must page or swap, a local disk gives the
- best performance.
-
-
- (2) Can you cache system routine pages on the server and thus improve
- performance of a sattelite node?
-
- No, because the system routine pages are cached on the free
- memory list of the sattelite node.
-
- You get best application performance by having "enough" memory
- to run it with minimal paging, and by tuning VMS to use all of
- that memory.
-
- Some experiments indicate that you can do quite well in LAVC performance
- improvements, without buying a single piece of additional hardware, by
- a few simple VMS tuning changes.
-
-
- Not that I would ever discourage you from buying more DEC hardware! :)
-
- ----------------------------------------------------------------------------
- | -Mike Pettigrew pettigrew_mi@ripple.enet.dec.com
- | (206) 661-3032
- | Digital Equipment Corporation
- | Federal Way, Washington, USA
- |___________________________________________________________________________
- | Surprises Every Day!
- |___________________________________________________________________________
- |These opinions are mine and not necessarily my employer's
- ----------------------------------------------------------------------------
-