Proceedings of the 23rd international conference on Parallel architectures and compilation

Abstract

It is an honor to introduce the technical program for the 23rd International Conference on Parallel Architectures and Compilation Techniques (PACT 2014). This symposium is one of the leading venues for new ideas and results in the area of parallel computing. This year's program includes 37 papers, 17 posters, keynotes from Klara Nahrstedt (University of Illinois) and Bob Blainey (IBM), a day of workshops and tutorials, and the ACM Student Research Competition. PACT 2014 received 144 paper submissions. I assigned each paper to at least 4 Program Committee (PC) members and 1 External Program Committee (EPC) member to review. To ensure the highest reviewing standards, the PC and EPC members were selected among the most recognized researchers in our field. Given that I had 51 PC members, each PC member had to review 11-12 papers personally. Overall, I believe that all of the PC and EPC members showed a very high degree of professionalism and fairness in their reviews. After all the reviews were collected, a Rebuttal Period allowed the authors to respond to the reviews. Then, PC and EPC members read the 5 reviews and the authors' response for the papers they had read, and engaged in a week-long discussion with other reviewers of the same paper(s) via email. At the end of this process, each PC and EPC member had to explicitly assign a grade to each of the papers she/he had reviewed. The papers' average grade was used to order the discussion of papers at the PC meeting. The whole review process was double blind. The PC meeting was held on May 10th, 2014, at the Chicago O'Hare Hilton. 46 PC members attended physically, while most of the others participated over the phone. The meeting started at 8am and lasted until 7:30pm. Before a paper was discussed, all conflicted PC members left the room and the authors' names were revealed. Then, the four PC members who had read the paper tried to reach a unanimous decision on the paper's outcome. If they could not, the whole PC voted. PC papers were discussed together with the other papers. The outcome of papers was not revealed to conflicted PC members until after the meeting. Over the course of the day, we discussed over 75 papers. We accepted 37 papers (25.7% acceptance rate) and 17 posters. To improve the quality of the program, several of the accepted papers were shepherded.