From gcf@boss.npac.syr.edu Wed Mar 11 09:18 EST 1998 Received: from boss.npac.syr.edu (boss.npac.syr.edu [128.230.117.20]) by postoffice.npac.syr.edu (8.7.5/8.7.1) with SMTP id JAA15033 for ; Wed, 11 Mar 1998 09:18:12 -0500 (EST) Received: from localhost (gcf@localhost) by boss.npac.syr.edu (950413.SGI.8.6.12/8.6.6) with SMTP id JAA16525 for ; Wed, 11 Mar 1998 09:19:56 -0500 Message-Id: <199803111419.JAA16525@boss.npac.syr.edu> X-Authentication-Warning: boss.npac.syr.edu: Host localhost didn't use HELO protocol Reply-to: gcf@npac.syr.edu To: kea@postoffice.npac.syr.edu Subject: CpandE Date: Wed, 11 Mar 1998 09:19:56 -0500 From: Geoffrey Fox Content-Type: text Content-Length: 3439 Status: RO C374 Letter We enclose 5 referee reports on: C374: As usual the referees do not present a totally uniform point of view but there are some clean common threads. Although we cannot publish your paper in its current form, we believe that a revised paper would be accepted as both the subject and approach seem interesting. The following two issues appear to be critical. Firstly is the paper up to date? Modern references and popular technologies such as UML should be discussed. Secondly the analysis is rather theoretical for our journal and needs to be more quantitative to exhibit the expected "Practice and Experience". We look forward to a modified paper, which should include a memo describing your response to the referees. We thank you for your interest in Concurrency:Practice and Experience. C375 Report This is an interesting paper but the motivating application to petaflops systems is inaccessible to most readers. Thus I suggest adding a short section describing the petaflops architecture and why the generalized hypercube emerges from a consideration of optical interconnects. This section will give an otherwise rather tough mathematical paper, the appropriate "Practice and Experience" needed by this journal. C375 Letter We enclose a referee report on C375: We would be pleased to publish this paper but would request the change specified by the referee. We look forward to a revised paper. We thank you for your interest in Concurrency:Practice and Experience. C376 Report 1 (Really Per Brinch Hansen) This paper is interesting but needs revisions before publication. It is not clearly stated that the basic algorithms are well known and the quoted work of Per Brinch Hansen predated this paper by several years. The paper should give more details on llc itself and de- emphasize the review of earlier sorting work. In particular issues such as virtual processors and dynamic process/memory management deserve more attention. As a more minor point, please explain what the difference between processor activation and processor assignment are and why this is connected to difference between llc and "hand- coded" message passing; I can hand code anything surely! Why does table 1 show "hand- coding" getting much better performance on 2 but poorer performance on 4-16 processors? C376 Report 2 (gcf) This paper contains interesting material but is largely focussed on one application -- sorting. llc can only be useful as a library if it is more generally applicable -- can you not discuss other ways of using llc? I wonder how the llc sorting implementation compares to those that do not use explicit recursion. Finally I find the figures of marginal quality and some could be left as tables. C376 Letter We enclose two referee reports on C376 The referees agree that the material is interesting and appropriate for the journal but we regret that we cannot publish the paper in its current form. However we believe it is clear how it can be modified and believe that a suitable revised version can be accepted without further review. We look forward to this revision which should contain a memo describing your response to the referees. We thank you for your interest in Concurrency:Practice and Experience. Geoffrey Fox gcf@npac.syr.edu, http://www.npac.syr.edu Director of NPAC and Professor of Physics and Computer Science Phone 3154432163 (Npac central 3154431723) Fax 3154434741