Proceedings of the ACM SIGMETRICS/international conference on Measurement and modeling of computer systems

Abstract

Welcome to SIGMETRICS 2013. SIGMETRICS is the flagship conference of the ACM special interest group for the computer systems performance evaluation community. This year marks the fortieth anniversary since SIGMETRICS (under its prior name, SIGME) held the First National SIGME Symposium on Measurement and Evaluation in 1973. The past four decades have seen enormous changes in the field of computer science, but the importance of measurement, modeling, and performance evaluation remains as critical as ever. This year's conference includes papers on topics that have been a mainstay since the founding of our SIG, including queuing, scheduling, resource allocation, and performance measurement. Application areas that have emerged in recent years, such as multicore systems, cellular networks, and energy optimization, continue to be represented in our program. Papers on solid-state storage have seen a significant uptick this year, and we have papers on some topics that are new to SIGMETRICS, including crowdsourcing and RFID systems. Interestingly, the program also shows a drop-off in topics that were hot just a brief while ago, such as social networks, BitTorrent, swarms, peer-to-peer, and MapReduce. We received 196 submissions to this year's conference, of which 26 appear in the program as full papers, which is a highly competitive acceptance ratio below 14%. An additional 28 submissions appear in the abbreviated form of poster presentations with brief summaries in the proceedings. As in some prior years, we performed reviews in two rounds. In the first round, each paper was assigned to four reviewers. In the second round, additional reviews were assigned to papers with fewer than three completed reviews and papers with highly divergent review opinions and fewer than two high-confidence reviews. We experimented with two changes to the review process this year. The first change to the review process was the addition of a rebuttal phase between the first and second review rounds, to give authors an opportunity to respond to questions raised in first-round reviews. To impede the addition of new substantive material in the rebuttals, and instead reserve rebuttals for merely highlighting information already contained in the submission, we strictly limited each rebuttal to 500 characters. It is not easy to gauge the effectiveness of the rebuttal process: There were many occasions during the PC meeting when reviewers commented on items in authors' rebuttals, which suggests that the rebuttals provided additional information; however, reviewers mostly found that their opinions were unchanged by what they read in the rebuttals. The second change to the review process was the use of rankings rather than ratings. Instead of rating their assigned papers with accept/reject recommendations, each PC member was asked to produce a list of assigned papers ordered by the reviewer's assessment of each paper's overall quality. Our intent was to eliminate the bias that is inherent in accept/reject recommendations because each reviewer has only a narrow view of the conference's submissions. Reviewers' individual rankings were combined into a global ranking using an algorithm similar to PageRank, and the top 60 papers were discussed at the PC meeting. During the PC meeting, whenever a paper was accepted, we identified any paper with global rank below 60 that at least one reviewer had ranked substantially higher than the accepted paper. The reviewer was given the option to add this other paper to the discussion list. About a dozen such additional papers were discussed, although none were accepted as full papers. We are pleased to present three awards to two of this year's papers. The SIGMETRICS Best Paper Award honors the overall best paper in each year's conference, and the Kenneth C. Sevcik Outstanding Student Paper Award honors an outstanding paper whose primary author is a student. This year, both awards are presented to an outstanding student paper that is also the overall best paper in the conference: "Queueing System Topologies with Limited Flexibility," by John N. Tsitsiklis and Kuang Xu. We are inaugurating a new award this year, the SIGMETRICS Best Practical Paper Award, to honor the best paper from among those whose research has the most direct practical applicability. This award is presented to "Practical Conflict Graphs for Dynamic Spectrum Distribution," by Xia Zhou, Zengbin Zhang, Gang Wang, Xiaoxiao Yu, Ben Y. Zhao, and Haitao Zheng.