home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!news.claremont.edu!ucivax!orion.oac.uci.edu!pamiri@uci.edu
- From: pamiri@uci.edu (Peter Amiri)
- Subject: Re: 4th Dimension Info Requested
- Nntp-Posting-Host: cam.oaa.uci.edu
- Message-ID: <2AA3F0D4.22355@orion.oac.uci.edu>
- Newsgroups: comp.sys.mac.databases
- Organization: Dimensional Software
- Lines: 23
- Date: 1 Sep 92 22:38:12 GMT
- References: <BtJDus.CEv@news.larc.nasa.gov> <1992Aug26.160438.12085@news.mentorg.com> <HUBBARD.92Aug31173527@tsc.cs.unc.edu>
-
- In article <HUBBARD.92Aug31173527@tsc.cs.unc.edu>, hubbard@tsc.cs.unc.edu (Robert Hubbard) writes:
- >
- >
- > We've had similar speed problems here at UNC-Chapel Hill Computer
- > Science Dept. In fact just switching from single-user to multi-user
- > modes in 4D seems to slow the performance down by a factor of 5.
-
- In single user mode 4D does data catching but when you switch to multi-user
- mode all catching is turned off and every thing has to be written back to
- disk. If you are on a network this can slow you done even further. The 4D
- server which should be very close to shipping now, should address this issue
- completely. 4D server will do data catching on the server so all request for
- data are handled in the most efficient way. Whats even more, if you have a
- output layout displaying a selection of records 4D server will only send
- across the fields on the output layout and not the entire records. This aloan
- will speed up things considerably.
-
- During both of the demos I've seen the server was on a PowerBook 170 using
- localtalk being accessed from another PB 170, IIci, and Quadra 700 and the
- speeds were impressive.
-
- Peter Amiri
- University of California, Irvine
-