NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/comp/dcom/modems/18783

Internet Message Format | 1993-01-02 | 3.6 KB

Xref: sparky comp.dcom.modems:18783 comp.sys.sun.hardware:6588 comp.sys.sun.admin:9864
Newsgroups: comp.dcom.modems,comp.sys.sun.hardware,comp.sys.sun.admin,cu.sun.managers
Path: sparky!uunet!zaphod.mps.ohio-state.edu!mstar!mstar!bob
From: [e-mail] (Bob Sutterfield)
Subject: Re: Anyone out there successfully get a 14.4K modem running on a Sun?
In-Reply-To: [e-mail]'s message of Fri, 1 Jan 1993 22: 37:58 GMT
Message-ID: <[e-mail]>
Sender: [e-mail]
Nntp-Posting-Host: roughy.morningstar.com
Organization: Morning Star Technologies
References: <[e-mail]>>
    <[e-mail]> <[e-mail]>
    <[e-mail]>
Date: Sat, 2 Jan 1993 06:59:03 GMT
Lines: 65

In article <[e-mail]> [e-mail] (Karl Denninger) writes:
   In article <[e-mail]> [e-mail] (Larry Snyder) writes:
      Have you played with other modems like the USR Courier v.32bis
      modems?  Folks (including myself) have found a reduced latency
      with these modems improving interactive throughput.

No, I haven't had the opportunity, but it's certainly possible that
other modems would add less latency than do the T3000s.  I've heard
reports to that effect, but I haven't tried them myself.

      Do you suggest MNP enabled on SLIP v.32bis lines -- or just
      v.32bis and v.42bis?

   Use V.42bis/LAPM!  MNP 5 can be a serious lose on compressed data!

Right.

   You're doing something wrong if your best text numbers are under
   3KB/sec (your TCP connection window is almost certainly too small).

Alas, that's something we can't adjust.  Our PPP looks to UNIX like an
IP interface (it shows up when you say "ifconfig -a"), and while we
can set the M[RT]U, we can't adjust TCP parameters.  We take whatever
the host UNIX provides.

   With proper window settings on both ends, and VJ header
   compression, I am getting ~35Kb/sec on text FTPs over a V.32bis
   link running PPP.

(You meant 3.5Kb/sec, right?)

   This is with >cheap< modems (Supras) on one end, and T3000s on the
   other.  If I force the Supras out of the loop so I'm running T3000
   <> T3000 I can achieve nearly 5Kb/sec >reliably< on most text files
   over 100K or so.

You must have the DTEs clocked at 57600, right?  The hardware I've
tested so far only goes 38400.  I'm looking forward to playing with
57600, 76800, and 115200; hopefully next week.

   This is with Netblazers on both ends, going to and from an Ethernet
   host on each side.  The ethernet propagation delay + host time is a
   few ms at most.

I was measuring between SPARCstations running PPP locally, through the
modems on their native serial ports, clocked at 38400.

   I would hope that Morningstar lets you tune these parameters
   (particularly window size and MSS); if not you're going to have to
   accept what you've got now.

Unfortunately, we're constrained by whatever window size the host UNIX
TCP/IP stack offers.  The only parameter we can affect is packet size.

   One word of warning -- setting these parameters up higher >can<
   affect Netblazer (or other router) memory usage severely as packet
   buffer memory consumption will rise dramatically.  This might be
   bad if you are tight on RAM.

Not a problem for us - our PPP/SLIP daemon runs in user space,
enjoying access to demand-paged virtual memory :-)
--
Bob Sutterfield, Morning Star Technologies            [phone]
1760 Zollinger Rd, Columbus Ohio USA, 43221           [phone]
[e-mail]                                   [phone] (FAX)