home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!haven.umd.edu!darwin.sura.net!zaphod.mps.ohio-state.edu!news.acns.nwu.edu!network.ucsd.edu!mvb.saic.com!macro32
- From: <brydon@dsny25.sinet.slb.com>
- Newsgroups: vmsnet.internals
- Subject: RE: VMS tuning for a vaxcluster (LAVC)
- Message-ID: <9208311910.AA15058@sndsu1.sinet.slb.com>
- Date: Mon, 31 Aug 92 19:10:38 GMT
- Organization: Macro32<==>Vmsnet.Internals Gateway
- X-Gateway-Source-Info: Mailing List
- Lines: 44
-
- Bob George (BOB@AENEAS.IMS.DISA.MIL) sez:
- >#Nope. The local disk's still faster. When I and a friend did VMS tuning
- >#studies for <a former employer> we found that even on an unloaded net the
- >#local disk was faster. Mind you, this was using VAXstation IIs with between
- >#5 and 16MB and RD54s. Mileage on newer systems may vary, but I'd bet that
- >#the local disk still beats Ethernet traffic.
- >
- >[lots of interesting stuff deleted]
- >
- >Am I missing something fairly obvious in this thread or on the nonlocal
- >paging, don't you still have to wait for a disk I/O to occur. On top of
- >the overhead of setting up the ethernet packet(s), when it gets to the
- >node where it was going, there is still a disk I/O that needs done. On
- >a page write, this activity could be buffered. But on a page read, the
- >process would still be hung up until the actual disk gets read and the data
- >returned. For at least moderately similar disks, this would give the
- >advantage to the local disks.
-
- To restate several points I was trying to make:
-
- (1) What is considered page/swap data on the satellite may (or may not?) be
- considered cache-able data on the boot node. Reads will probably show
- different numbers from writes any way you look at it. In a previous message
- in this thread, somebody (from DEC?) indicated that there could be no caching
- on either reads or writes for some reason...what is the explanation for that?
-
- (2) Yes it takes ethernet processing, but even on a slow VMS system, this will
- be on the order of a few microseconds. Okay, say it is 100 microseconds. One
- millisecond of disk controller latency (or any other delay) is 1000
- microseconds. Access times on disks are typically 20000 to 50000 microseconds
- (remember the difference between 'average access time' and 'maximum access
- time' and 'typical access time').
-
- (3) In many cases, the hardware technology on the boot node is much better
- (eg. faster) than that on the satellite node. This goes back to the birth of
- the technology with 45 msec RD53/RD54s on VAXstation 2000's and 25 msec DSA
- drives on an 86xx boot node. Assume 1 millisecond for ethernet overhead and
- processing. With this configuration, 25 + 1 is much less than 45. The access
- time numbers for boot nodes versus satellites is getting closer but a
- difference (based on technology) still seems to be the case.
- _______________________________________________________________
- Harvey Brydon | Internet: brydon@dsn.SINet.slb.com
- Dowell Schlumberger | P.O.T.S.: (918)250-4312
- Reality is for those that can't handle simulation.
-