ram@nucsrl.UUCP (04/17/87)
I am curious to know about:
Has anybody tried NFS over long haul networks? (Something like ARPA,
dedicated High speed links over large geographical distances ....).
This arose out of this evening's fantasy discussion with my friend
(whose name I shall not reveal as he might be too embarassed to
publicize his fantasies and affections for NFS, UNIX ....). We
were fantasizing about a global file access system (yes, mother Earth)
Will it be a fantasy for ever? I guess so.
UNIXes of the world unite.
[Imagine the paranoia at NSA, MD. gorby@kgbvax can 'cd' into
/earth/us/md/nsa/secret_files/ at nsacray!].
-------------------
Renu Raman UUCP:...ihnp4!nucsrl!ram
1640 Balmoral Circle ARPA:ram@eecs.nwu.edu (Soon. Not Yet)
Palatine IL 60067 AT&T:(312)-869-4276
P.S: As I typed this note initially, I got a message "note file cannot be
created: possibly archives going on" - words to that effect. Guess the
NSA is running a note kill daemon on this machine. My determination has
not waned.cyrus@hi.UUCP (04/19/87)
In article <3680010@nucsrl.UUCP> ram@nucsrl.UUCP (Renu Raman) writes: > Has anybody tried NFS over long haul networks? (Something like ARPA, > dedicated High speed links over large geographical distances ....). Here at the University of New Mexico we soon will have that capability. We are getting a T1 line (~1.4Mbits/sec) put in between the U and Los Alamos National Labs. Although we will NOT be running NFS on it we will be running tcp/ip (some of the time). It will be primarily used for distributed simulations; i.e. message passing. -- @__________@ W. Tait Cyrus (505) 277-0806 /| /| University of New Mexico / | / | Dept of EECE - Hypercube Project @__|_______@ | Albuquerque, New Mexico 87131 | | | | | | hc | | e-mail: | @.......|..@ cyrus@hc.dspo.gov or cyrus@hc.arpa or | / | / {gatech|ucbvax|convex}!unmvax!hi!cyrus @/_________@/
wesommer@athena.mit.edu (William Sommerfeld) (04/20/87)
In article <3680010@nucsrl.UUCP> ram@nucsrl.UUCP (Renu Raman) writes: > Has anybody tried NFS over long haul networks? (Something like ARPA, > dedicated High speed links over large geographical distances....). Yes. It doesn't work very well over a long haul ARPA link (say, MIT->UCB). It probably works OK over T1 links (like the NSFnet backbone). NFS was designed for high-speed, low-latency, high-reliability nets like Ethernet. On long-haul nets, two things kill NFS: the round trip time, and the maximum packet size. Every pathname *component* lookup takes one exchange with the server. On a local ethernet, this takes on the order of a few milliseconds (~10). Over the ARPAnet, which is made of 56KB links for the long haul, 1 *second* is more like reality. This means that looking up, say /usr/spool/news/net/unix/wizards/400 on UCBVAX would take 7 roundtrips, probably taking seven or eight seconds. Then, after that, each read, of no more than perhaps 450 bytes, would take another second. (NFS returns about 70 bytes of file statistics information with each read request, given that the maximum packet size on the ARPAnet is 576 bytes, and the UDP/IP headers themselves take perhaps 40 bytes). That's a throughput of 3.2KBits/second on a 56KB/sec link - about 6% of the bandwidth theoretically possible. According to SUN, NFS over ethernet between a pair of SUN-3's moves more like 250 KBytes/sec (2MBit/sec), which is 20% of Ethernet speed. Bill Sommerfeld MIT Project Athena ARPA: wesommer@athena.mit.edu UUCP: ...!mit-eddie!wesommer
hays@apollo.uucp (John Hays) (04/21/87)
In article <4597@hi.uucp> cyrus@hc.dspo.gov (Tait Cyrus) writes: >In article <3680010@nucsrl.UUCP> ram@nucsrl.UUCP (Renu Raman) writes: >> Has anybody tried NFS over long haul networks? (Something like ARPA, >> dedicated High speed links over large geographical distances ....). > >Here at the University of New Mexico we soon will have that capability. >We are getting a T1 line (~1.4Mbits/sec) put in between the U and >Los Alamos National Labs. Although we will NOT be running NFS on it Domain Distributed Services (part of Apollo's implementation of U*ix) has an NFS capability and DDS runs over T1 links quit satisfactorally --- Here at our offices we use a 60 mi T1 to do day to day Demand Paged Virtual Memory between systems in Mass. and New Hampshire. It is kind of fun, however it is so transparent you forget that it is there. John -- John D. Hays, Consultant UUCP: ...!decvax!wanginst!apollo!hays Corporate Systems Engineering ...!uw-beaver!apollo!hays Apollo Computer Inc. CIS: 72725,424 {weekly} !MY OPINIONS, not Apollo's!
lindsay@cheviot.UUCP (04/24/87)
In article <3680010@nucsrl.UUCP> ram@nucsrl.UUCP (Renu Raman) writes: > > UNIXes of the world unite. > This is a registered slogan of the Newcastle Connection - please do not mention it in the same breath as the *%@&#*!~@%#$!@* NFS!!!!!! (See Software Practice and Experience Dec 1982) -- Lindsay F. Marshall JANET: lindsay@uk.ac.newcastle.cheviot ARPA: lindsay%cheviot.newcastle@ucl-cs PHONE: +44-91-2329233 UUCP: <UK>!ukc!cheviot!lindsay "How can you be in two places at once when you're not anywhere at all?"