************************************************************************

       ATM Forum Document Number: ATM Forum/94-1173

       ************************************************************************

       Title:  Transient Performance of EPRCA and EPRCA++

       ************************************************************************

       Abstract:
       Transient performance of EPRCA and EPRCA++ (a modification of EPRCA+)
       is compared.

       ************************************************************************

       Source:

       Raj Jain, Shiv Kalyanaraman and Ram Viswanathan
       The Ohio State University
       Department of CIS
       Columbus, OH 43210-1277
       Phone: 614-292-3989, Fax: 614-292-2911, Email: Jain@ACM.Org

       The presentation of this contribution at the ATM Forum is sponsored by
       NIST.

       ************************************************************************

       Date:   November 29-Dec 2, 1994

       ************************************************************************

       Distribution: ATM Forum Technical Working Group Members
                     (Traffic Management)

       ************************************************************************

       Notice: This contribution has been prepared to assist the ATM
       Forum.  It is offered to the Forum as a basis for discussion and
       is not a binding proposal on the part of any of the contributing
       organizations.  The statements are subject to change in form and
       content after further study.  Specifically, the contributors
       reserve the right to add to, amend or modify the statements
       contained herein.

       ************************************************************************

       The transient performance of EPRCA and many of its variation is
       rather slow. It takes several milliseconds to get to the steady
       state. In real world, the networks are always operating in a
       transient mode. Some VCs are finishing up while others are
       starting up. Transient performance is, therefore, an important
       aspect. On LAN systems, where round-trip delay is only a few
       microsecons, EPRCA takes several milliseconds to reach the steady
       state.  Graphs of the simulation results will be presented at the
       meeting.

       EPRCA++ is a modification of EPRCA+ presented at the October
       interim meeting.  The key objections to EPRCA+ was its use of
       time-based RM cell transmission.  This is replaced by a count-
       based RM cell transmission. All other benefits of EPRCA+
       including the following  apply to EPRCA++:

       1. Congestion avoidance (small queue length),
       2. Simplicity,
       3. Small number of parameters,
       4. Insensitivity to parameters, and
       5. Easy to set parameters

       In addition, the following are new features of EPRCA++:

       1. The queue length goal can be set at any desired value
       2. EPRCA++ can operate at any desired operating point between the
       knee and the cliff of the delay-throughput curve.
       3. 100%  link utilization is achieved
       4. VCs can start at PCR (ICR=PCR). However, the impact on the
       performance is minimal since the queue lengths are controlled.
       The maximum queue length is equal to the theoretical maximum (one
       round trip delay) and  come down fast. At least in LANs, the
       impact of VCs starting at PCR is minimal.
       5. Idle VCs are not required to do any computation during or
       after the idle period.

       Simulation results comparing EPRCA and EPRCA++ will be presented.

       [1] Raj Jain, Shiv Kalyanaraman, and Ram Viswanathan, "Simulation
       Results: EPRCA+ Scheme," ATM Forum Traffic Management Group
       Interim Meeting, October 25-26, 1994.