Subject: Re: Request to review a paper C516 for Concurrency & Computation: Practice and Experience From: "Nikos P. Chrisochoides" Date: Wed, 15 Aug 2001 10:13:21 -0400 To: gcf@indiana.edu CC: nikos@cs.wm.edu X-UIDL: 96d74aae33130000 X-Mozilla-Status: 0011 X-Mozilla-Status2: 00000000 Received: by mailer.csit.fsu.edu (mbox gcfpc) (with Cubic Circle's cucipop (v1.31 1998/05/13) Wed Aug 15 10:29:34 2001) X-From_: fox@mailer.csit.fsu.edu Wed Aug 15 10:15:52 2001 Return-Path: Delivered-To: gcfpc@csit.fsu.edu Received: from dirac.csit.fsu.edu (dirac.csit.fsu.edu [144.174.128.44]) by mailer.csit.fsu.edu (Postfix) with ESMTP id 3E65923A13 for ; Wed, 15 Aug 2001 10:15:52 -0400 (EDT) Received: from localhost by dirac.csit.fsu.edu (AIX4.2/UCB 8.7) id KAA120196; Wed, 15 Aug 2001 10:15:51 -0400 (EDT) Resent-Message-Id: <200108151415.KAA120196@dirac.csit.fsu.edu> Delivered-To: fox@csit.fsu.edu Received: from fins.uits.indiana.edu (fins.uits.indiana.edu [129.79.6.185]) by mailer.csit.fsu.edu (Postfix) with ESMTP id 592B223A13 for ; Wed, 15 Aug 2001 10:13:23 -0400 (EDT) Received: from va.cs.wm.edu (va.cs.wm.edu [128.239.2.31]) by fins.uits.indiana.edu (8.10.1/8.10.1/IUPO) with ESMTP id f7FEDMC02337 for ; Wed, 15 Aug 2001 09:13:22 -0500 (EST) Received: from in.cs.wm.edu (in [128.239.2.21]) by va.cs.wm.edu (8.11.4/8.9.1) with ESMTP id f7FEDLm28638; Wed, 15 Aug 2001 10:13:21 -0400 (EDT) Received: (from nikos@localhost) by in.cs.wm.edu (8.11.2/8.9.1) id f7FEDLU04313; Wed, 15 Aug 2001 10:13:21 -0400 Message-Id: <200108151413.f7FEDLU04313@in.cs.wm.edu> X-Authentication-Warning: in.cs.wm.edu: nikos set sender to nikos@in.cs.wm.edu using -f In-reply-to: <3B4F7414.9060901@csit.fsu.edu> (message from Geoffrey Fox on Fri, 13 Jul 2001 17:20:04 -0500) References: <3B4F7414.9060901@csit.fsu.edu> User-Agent: SEMI/1.13.7 (Awazu) CLIME/1.13.6 (???) Emacs/20.7 (i386-redhat-linux-gnu) MULE/4.0 (HANANOEN) MIME-Version: 1.0 (generated by SEMI 1.13.7 - "Awazu") Content-Type: text/plain; charset=US-ASCII Resent-To: Geoffrey Fox Resent-Date: Wed, 15 Aug 2001 10:15:51 -0400 Resent-From: Geoffrey Fox Dear Geoffrey, I append the reviews from the C516 paper. Overall this is a good short paper on AMs implementation on top of VIA. C: Paper and Referee Metadata * Paper Number: C516 * Date: August 14, 2001 * Paper Title: An Analysis of VI Architecture Primitives in Support of Parallel and Distributed Communication * Author(s): Andrew Begel, Philip Buonadonna, David E. Culler, David Gay University of California, Berkeley * Referee: Nikos Chrisochoides * Address: Collge of William and Mary Referee Recommendations. Please indicate overall recommendations here, and details in following sections. 2. accepted provided changes suggested are made D: Referee Comments (For Editor Only) This paper describes the implementation of AMs on the VI Architecture. The implementation section is very brief and the comparison with two previous implementations, too short. Instead the authors choose to spend half of the paper to related work, VIA, Infiniband, AMs, and Split-C, widely known and well published systems. Although, it is important for completeness to introduce the reader to all these systems, the first half of the paper can be reduced in half. Instead more details --as it is claimed in line 1 of the Abstract-- can be given on the implementation and evaluation of AMs/Split-C on VIA. I do not have to see the re-visions before the publication of the paper. E: Referee Comments (For Author and Editor) This is a good systems implementation paper. The main contributions of the paper are two: (1) a description of mapping AMs and Split-C abstractions on VIA, and (2) an evaluation of the implementation ---the cost of implementing AMs on top of VIA primitives is analyzed in terms of the cost of VIA primitives. In addition the authors use this experience to discuss implications for Infiniband and future high-performance network architectures. Sections 5 and 6 are very short. VIA is widely available and accessible to many researchers and parallel programmers unlike the NIC software of other parallel machines on which AMs are implemented. The paper will be very useful with great impact if the authors describe in more detail the implementation of AMs functions on VIA. This paper along with the code could be of great development value and excellent educational tool for parallel programmers. F: Presentation Changes Remove Figure 3 and 4. They are redundant and hard to read, especially the legends. .