simulation@uflorida.cis.ufl.edu (Moderator: Paul Fishwick) (12/01/89)
Volume: 12, Issue: 9, Fri Dec 1 10:23:49 EST 1989 +----------------+ | TODAY'S TOPICS | +----------------+ (1) Reducing Time Warp Overhead (2) Call for Papers: DEVS & Advanced Modelling (3) Time Warp (4) Knowledge Based Simulation of Circuits * Moderator: Paul Fishwick, Univ. of Florida * Send topical mail to: simulation@bikini.cis.ufl.edu OR post to comp.simulation via USENET * Archives available via FTP to bikini.cis.ufl.edu, login as 'ftp', use your last name as the password, change directory to pub/simdigest. * Simulation Tools available by doing above and changing the directory to pub/simdigest/tools. ----------------------------------------------------------------------------- Date: Wed, 22 Nov 89 17:12:11 -0800 From: liny@cs.washington.edu (Yi-Bing Lin) Return-Path: <liny> To: fishwick@bikini.cis.ufl.edu Subject: Reducing the state saving overhead of Time Warp The following technical report (TR 89-09-04) can be requested via e-mail: tech-report@june.cs.washington.edu THE OPTIMAL CHECKPOINT INTERVAL IN TIME WARP PARALLEL SIMULATION Yi-Bing Lin and Edward D. Lazowska Department of Computer Science and Engineering University of Washington Seattle, WA 98195 ABSTRACT The Time Warp mechanism is the most common ``optimistic'' parallel simulation protocol. A process executes every message as soon as it arrives. If a message with a smaller timestamp subsequently arrives, the process rolls back its state to the time of the earlier message and re-executes from that point. Clearly, the state of each process must be saved (checkpointed) regularly in case rollback is necessary. Although most existing Time Warp implementations checkpoint after every state transition, this is not necessary, and the checkpoint interval is in reality a tuning parameter of the simulation. This paper shows how to derive the optimal frequency of checkpointing in Time Warp simulation. We express the optimal checkpoint interval as a function of the average rollback distance, the average state saving overhead, the ratio of the total number of executed messages to the total number of undone messages, and the average cost of a simulation step. The result is consistent with the following intuition: A large checkpoint interval should be chosen if and only if the state saving overhead is large, and/or a large number of events are executed, on the average, between two consecutive rollbacks. ------------------------------ Date: Thu, 30 Nov 89 13:58 CDT From: TKIM@kuhub.cc.ukans.edu Subject: Ad for a call for papers To: fishwick@fish.cis.ufl.edu X-Vms-To: @fish Call for papers SIMULATION DIGEST's Special issue on "The DEVS Formalism and Advanced Modelling Simulation Methodology" Authors are invited to submit abstracts of papers describing recent advances on all aspects of the DEVS (Discrete Event System Specification) formalism. These may include simulation environments implementing the formalism and applications of the environments to modelling and simulation of various discrete event systems. Abstracts should be not more than two double-spaced pages. Abstracts must be received not later than Jan 15 of 1990. Accepted papers will be determined by Jan 31 of 1990. Accepted full papers will be due by March 15 of 1990 in order to be published in the special issue of the SIMULATION DIGEST for 1990. Mail abstracts to: Prof. Tag Gon Kim Dept of Electrical and Computer Engineering The University of Kansas Lawrence, KS 66045 e-mail: "tkim@ukanvax" (ph) 914-864-7740 --------------------------------------------------------------------------- From: "cp.jobling eleceng staff" <eechris%pyramid.swansea.ac.uk@NSFnet-Relay.AC.UK> Date: Thu, 30 Nov 89 20:40:18 GMT To: simulation@bikini.cis.ufl.edu Hi, I've just become aware that 'time-warp' may be the answer to something I want to do in object oriented techniques for continuous systems simulation. I am therefore looking for pointers to key references, preferably those that are easy to get hold of. Is there anybody out there who could mail me a bibliography? The only references I've come across so far are a couple of tech-reports from the Rand Corporation. Maybe someone from that body could tell me how to get hold of these? Thanks Chris ======================================================================== Chris Jobling, Department of Electrical and Electronic Engineering, University College, Swansea, Singleton Park, Swansea SA2 8PP, UK ------------------------------------------------------------------------ Phone: +44-792-295580 Fax: +44-792-295532 Telex: 485358 email: JANET: eechris@uk.ac.swan.pyr Internet eechris%pyr.swan.ac.uk@cunyvm.cuny.edu Earn/Bitnet eechris%pyr.swan.ac.uk@UKACRL UUCP eechris%pyr.swan.ac.uk@ukc.uucp EAN eechris@pyr.swan.ac.uk ======================================================================== ------------------------------ From: Klaus Milzner <milzner@unidocv.informatik.uni-dortmund.de> Date: Fri, 1 Dec 89 14:44:55 +0100 To: simulation@bikini.cis.ufl.edu Hi out there ! I currently do some research in the area of circuit specific, knowledge based simulation support. The aim is the realization of a simulation assistant capable of automatically characterizing certain classes of circuits and providing estimations of the simulation and design results. Though simulation is a central aspect of nearly every design cycle, surprisingly little attention seems to be payed to this topic in the past. To complete my studies of literature, i'm urgently looking for any pointers, references etc. concerning this field. ANY comments are appreciated...:-) Thanks in advance, Klaus PS: Are there any information about NECTAR from UCB available ? +-----------------------------------------------------------------------------+ | Klaus Milzner University of Dortmund Lehrstuhl Informatik 1 | | milzner@unidocv.uucp | +-----------------------------------------------------------------------------+ ------------------------------ END OF SIMULATION DIGEST ************************