************************************************************************ ATM Forum Document Number: ATM Forum/95-1344 ************************************************************************ Title: New Source Rules and Satellite links ************************************************************************ Abstract: The effect of new CIF and ICR formulae on satellite networks is studied. Source rule 5 (TOF dectease) seems to have mostly negative effect on performance. Rescheduling, on the other hand, has positive effect. ************************************************************************ Source: Raj Jain, Shiv Kalyanaraman, Fang Lu, and Sonia Fahmy The Ohio State University Department of CIS Columbus, OH 43210-1277 Phone: 614-292-3989, Fax: 614-292-2911, Email: Jain@ACM.Org Saragur Srinidhi NASA Lewis Research Center and Sterling Software Scientific Systems Division 21000 Brookpark Road, MS 141-1 Cleveland,OH 44135 Phone: 216-433-8987, Fax: 216-433-8000 The presentation of this contribution at the ATM Forum is sponsored by NASA. ************************************************************************ Date: October 1995, Honolulu ************************************************************************ 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. ************************************************************************ In this contribution, we intend to present the effect of the following three rules on satellet links: 1. CIF and RTT are used to determine XRM and ICR. This change was introduced in the August meeting. We have studied the effect of this change and noticed that this sometimes causes problems that we did not anticipate. 2. Source Rule 5 has been found to not achieve its intended effect. It causes oscillations in long delay links such as in satellite networks. The reason of oscillation is that the low rate of the source results in timeouts even if the source keeps requiring to increase in rate. Rule 5 bring the rate down shortly after the source rate get increased, so the oscillation occurs. 3. Rescheduling was accepted as an option [1] and introduced as a minor change to pseudocode in section I.1, pg. 87. We have studied the effect of rescheduling and found that it allows fast response and throughput improvements (as opposed to ACR increase) for sources whose rates are set to low values, and in many cases avoids unnecessary triggering of conditions like Tof, Trm and Xrm, due to inactivity. It speeds up slow reverse traffic and improves the inter BRM cell time, and hence the transient response is speeded up. We substantiate these points with simulation results in an accompanying contribution. Rescheduling also helps protect unnecessary triggering of Rule 5. Without rescheduling, we found that in some cases, the rule 5 is triggered resulting in oscillations in source rate. With rescheduling, Rule 5 is not triggered so that the network is stable and the throughput is improved. This happens because the reverse traffic is speeded up so that the interval between BRM is shorter and more even and the source may receive the feedback faster, avoiding the trigger of Tof, Trm and Xrm to bring the source rate down and avoiding oscillation. Note: All our contributions and slides are available through our web site: http://www.cse.wustl.edu/~jain/atmforum.htm