home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.vms
- Path: sparky!uunet!haven.umd.edu!decuac!pa.dec.com!engage.pko.dec.com!nntpd.lkg.dec.com!star.enet.dec.com!buda
- From: buda@star.enet.dec.com (Mark A. Buda)
- Subject: Re: Striping and Shadowing
- Message-ID: <1993Jan22.155201.29994@nntpd.lkg.dec.com>
- Sender: usenet@nntpd.lkg.dec.com (USENET News System)
- Organization: Digital Equipment Corporation
- Date: Fri, 22 Jan 1993 15:48:00 GMT
- Lines: 37
-
-
- In article <76FB3BE080600321@THINK2.LakeheadU.Ca>, OCCHILL@ROWDI.LakeheadU.Ca (Geoff Hill) writes...
-
- > 2) Volume Shadowing:
- > - how effective in improving disk I/O ?
-
- I will let someone else answer the other questions, but want to
- help share information about volume shadowing improving I/O.
-
- Volume shadowing is not designed and meant to improve disk I/O.
-
- Volume shadowing is a data availability product. It can help improve
- performance (read), but you need to plan worst case. At DECUS and other
- forums people talk about and some buy it for the possibility of an
- increase in performance - but DEC does not warranty the software for
- this function.
-
- EXAMPLE:
-
- You have a shadow set with 2 members. While both members are available,
- you might see a performance increase. This is great! During your
- heaviest time of access, end of month, the HDA on one of the members
- goes belly up! You are now left with a ONE member shadow set. Your
- performance gain that you had with two is no longer available. You find
- the month end might not finish in time because only one disk drive is
- available to service requests.
-
- Plan for WORST case and if you happen to get better performance in
- the normal case - GREAT!
-
- Hope this helps,
-
- - mark
-
- buda@star.enet.dec.com
- ...!decwrl!star.enet.dec.com!buda
- buda%star.enet.dec.com
-