Cell Relay Archive

Cell Relay Retreat>List Archive>month:1999-Jan> msg00167



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Re: CR: Re: relationship bewteen PCR, SCR and MBS

  • From: "ronald h. davis" <ronaldd@lucent.com>
  • Date: Thu, 28 Jan 1999 11:41:57 -0600
  • Cc: cell-relay@cell.onecall.net
  • Organization: Lucent Technologies
  • Original-CC: cell-relay@cell.ucs.indiana.edu

Jan Novak wrote:
> 
> A basic rule for MBS is to be at least MTU of the media, behind ATM
> connection (ethernet in LANE for example e.g. about 35 cells)
> or AAL5 MTU if you have native ATM connection to the endsystem.
>

but you really do have to know something about the traffic pattern.
for instance, with data traffic, packet bursts may have different
patterns for a host, versus a server farm.  or, the traffic source may
be a media stream, such as video, in which rapidly changing, high 
action scenes can generate large bursts of traffic while most scenes 
may vary little from frame to frame.

-- 
  __  ______  __  / __/ |       lucent technologies, naperville il, usa
_/ (_(_) / (_(_/_/_(_/  .       ronald.h.davis@lucent.com
"if you ain't the lead dog, the scenery never changes"  -- edmund wilson