The Journal of Instruction-Level Parallelism
1st JILP Workshop on Computer Architecture Competitions (JWAC-1):

Cache Replacement Championship

in conjunction with:
ISCA-37  http://isca2010.inria.fr/

The workshop on computer architecture competitions is a forum for holding competitions to evaluate computer architecture research topics. The first workshop is organized around a competition for cache replacement algorithms. The Cache Replacement Championship (CRC) invites contestants to submit their replacement algorithm code to participate in this competition. Contestants will be given a fixed storage budget to implement their best replacement algorithms on a common evaluation framework provided by the organizing committee.

Objective

The goal for this competition is to compare different cache replacement algorithms for a last level cache in a common framework. Replacement algorithms will be evaluated for both private and shared last level caches.  The algorithms must be implemented within a fixed storage budget as specified in the competition rules. Submissions will be evaluated based on their performance using the framework provided by the organizing committee. Submissions will be evaluated for two configurations: a single-core configuration with a 1 MB last level cache, and a 4-core configuration with a 4 MB shared last level cache.

 

Prizes

The championship has two tracks: A single-core track and a multi-core track. The top performer for each track will receive a trophy commemorating his/her triumph (OR some other prize to be determined later). Top submissions will be invited to present at the workshop, when results will be announced. All source code, write-ups and performance results will be made publicly available through the CRC website. Authors of accepted workshop papers will be invited to submit full papers for possible inclusion in a special issue of the Journal of Instruction-Level Parallelism (JILP).  Inclusion in the special issue will depend on the outcome of JILP's peer-review process: invited papers will be held to the same standard as regular submissions.

 

Submission Requirements

Each contestant is allowed a maximum of three submissions to the competition. Each submission should include the following:

o     Abstract: A 300-word abstract summarizing the submission. In addition, the abstract should include the author names, their affiliations, and the email address of the contact author.

o     Paper: This will be a conference-quality write-up of the replacement algorithm, including references to relevant related work. The paper must clearly describe how the algorithm works, how it is practical to implement, and how it conforms to the contest rules and fits within the allocated storage budget. The paper must be written in English and formatted as follows: no more than four pages, single-spaced, two-column format, minimum 10pt Times New Roman font. The paper should be submitted in .pdf format, and should be printable on letter-size paper with one-inch margins on all sides. A submission will be disqualified if the paper does not clearly describe the algorithm that corresponds to the submitted code. Papers that do not conform to the length and format rules will only be reviewed at the discretion of the program committee. If a contestant has two or three similar submissions (e.g., only differ in table sizes or little tweaks to the algorithm), he/she should submit only one paper for all submissions. However, if the two or three submissions by one contestant are totally different, he/she should submit a paper for each submission.

o     Cache Replacement code: A single C++ header file that can be included in the provided infrastructure must be submitted along with the paper as a separate file. This code must be well commented so that it can be understood and evaluated. Unreadable or insufficiently documented code will be rejected by the program committee. The file sample_cacherepl.h from the infrastructure is distributed with a default true LRU replacement algorithm, and should be replaced with the contestant’s code. The code should be compiled and run on the existing infrastructure without changing any code or Makefile, and should NOT require any library code that is not part of C++.

More details on how to submit these files will be available before the submission deadline.

 

 

Competition Rules

 

The competition will proceed as follows. Contestants are responsible for implementing and evaluating their algorithm in the distributed framework. The framework itself is provided as a binary and cannot be modified (except for the header file that implements the cache replacement algorithm). Submissions will be taken, compiled and run with the original version of the framework. The contestants will be ranked on the basis of the measured performance of their replacement algorithms. Each contestant will get a score for each of the two configurations evaluated in the competition. In the single-core track, the score represents the geometric mean of their replacement algorithm speedups across a set of (undistributed) benchmarks. In the multi-core track, the score represents the geometric mean of the replacement algorithms weighted speedup across a set of (undistributed) multi-program and multi-threaded benchmarks. The distributed simulation framework would enable contestants to build their own benchmark files for use with the simulator.

 

 

Acceptance Criteria

 

In the interest of assembling a quality program for workshop attendees and future readers, there will be an overall selection process, of which performance ranking is the primary component. To be considered, submissions must conform to the submission requirements described above. Submissions will be selected to appear in the workshop on the basis of the performance ranking, novelty, and overall quality of the paper and commented code. Novelty is not a strict requirement, for example, a contestant may submit his/her previously published design or make incremental enhancements to previously proposed design. In such cases, performance is a heavily weighted criterion, as is overall quality of the paper (for example, analysis of new results on the common framework, etc.).

 

Description of the Simulation Infrastructure

 

CRC Kit: Download and Directions

 

Submission Instructions

 

Important Dates

 

Competition formally announced:

February 1, 2010

Evaluation framework available:

February 5, 2010

Submissions due:

April 16, 2010

Acceptance notification:

May 7, 2010

Final version due:

June 4, 2010

Results announced:

at workshop (June 20, 2010)

 

Workshop Program

 

Steering Committee

Alaa R. Alameldeen, Intel

Eric Rotenberg, NC State

 

 

 

 

Organizing Committee

Alaa R. Alameldeen, Intel

Aamer Jaleel, Intel

Moinuddin Qureshi, IBM

 

 

 

 

Program Chair

Joel Emer, Intel

 

 

 

Program Committee

Doug Burger, Microsoft

Mainak Chaudhuri, IITK

Aamer Jaleel, Intel

Gabriel Loh, Georgia Tech

Moinuddin Qureshi, IBM

Yan Solihin, NC State

 

Affiliated Logos

 

 

 


                                                  JILP