home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.protocols.nfs:2277 comp.protocols.tcp-ip.ibmpc:5143 comp.dcom.lans.ethernet:1867
- Newsgroups: comp.protocols.nfs,comp.protocols.tcp-ip.ibmpc,comp.dcom.lans.ethernet
- Path: sparky!uunet!sun-barr!ames!Mongo-Jr.Empirical.COM!karl
- From: karl@empirical.com (Karl Auerbach)
- Subject: Re: Sun <=> PC Transfer Rate (Summary)
- Message-ID: <karl.23.716089551@empirical.com>
- Sender: usenet@news.arc.nasa.gov
- Organization: Empirical Tools and Technologies
- References: <1992Sep4.173931.24033@pixel.kodak.com>
- Date: Thu, 10 Sep 1992 01:45:51 GMT
- Lines: 72
-
- >From: tgl@ssd.kodak.com (Tom Lathrop)
- >Subject: Sun <=> PC Transfer Rate (Summary)
- >Date: 4 Sep 92 17:39:31 GMT
- >I was asked to summarize the responses to my query about optimizing
- >Sun to PC file transfer.
-
- >>I would like to find a fast way to transfer large files from a Sun
- >>to a 486 PC.
-
- >The consensus seems to be that FTP is faster than NFS for file transfer.
- >FTP is included with the Sun PC-NFS package
-
- That is almost always the case. But it will be strongly affected by the
- window size parameters at either end.
-
- Was UDP checksumming turned on with NFS?
-
- Was FTP doing binary or text mode transfer? The latter requires the CPU at
- either end to examine each and every byte and do some data juggling. (
- Actually the PC end can win at this game because the DOS end-of-line
- pair is the same as the internet end-of-line.)
-
- And if you use TCP with the new window extension mechanisms you can get
- much better throughput. However, I don't know any PC code that uses those
- yet.
-
- There is also a substantial difference depending on the latency and error
- rate of the network. If you are going through routers and over the internet
- you can see a major difference between NFS and TCP file transfer rates.
-
- But the #1 absolute most important factor is the speed of the disk.
-
- I routinely get 2megabit FTP transfers from a Sun Sparc 2 to a wimpy
- 16mhz 386sx box with an 8 bit WD board and packet driver.... as long as the
- destination is NUL: (Software is PC/TCP v2.1)
-
- I've gotten as high as 5megabits/second using a 50Mhz 486DX box with a 16
- bit WD board using the same software and packet driver.
-
- (This was done on an Ethernet with a background load of 300+ packets/second.)
-
- But throw a disk in... and things get slower. Of course this is affected
- by how much disk caching is going on at either end.
-
- Two
- >SPARCstation 2's are apparently capable of data rates of 800 to 1000
- >kilobytes per second memory-to-memory transfer on a dedicated Ethernet.
-
- Sun's can do *much* better than that.
-
-
- > The CPU is probably not critical,
-
- CPU *is* critical. One of the big bottlenecks is checksum calculation.
- (I do hope that whoever measured the Sun NFS speed did it with checksums
- turned on -- Sun workstations run with UDP checksums OFF by default.)
-
- Also, some drivers assume 8086 instruction set and don't use the much faster
- block-move to I/O port instructions.
-
- In fact some drivers use DMA although a 386 or 486 CPU is often faster.
-
- >software side, a couple of people said that if a packet driver is used,
- >which one it is makes a big difference.
-
- I've sent over 2500 packets per second through packet drivers. And that
- was on a 12mhz 286. Of course, that was a special test designed to
- load the network.
-
- I routinely receive a load of over 500 packets/second with a packet driver.
-
- --karl--
-