ACM Home Page
Please provide us with feedback. Feedback
QuakeTM: parallelizing a complex sequential application using transactional memory
Full text PdfPdf (589 KB)
Source
International Conference on Supercomputing archive
Proceedings of the 23rd international conference on Supercomputing table of contents
Yorktown Heights, NY, USA
SESSION: Transactional memory I table of contents
Pages 126-135  
Year of Publication: 2009
ISBN:978-1-60558-498-0
Authors
Vladimir Gajinov  Barcelona Supercomputing Center, Barcelona, Spain
Ferad Zyulkyarov  Barcelona Supercomputing Center, Barcelona, Spain
Osman S. Unsal  Barcelona Supercomputing Center, Barcelona, Spain
Adrian Cristal  Barcelona Supercomputing Center, Barcelona, Spain
Eduard Ayguade  Barcelona Supercomputing Center, Barcelona, Spain
Tim Harris  Microsoft Research Cambridge, Cambridge, United Kingdom
Mateo Valero  Barcelona Supercomputing Center, Barcelona, Spain
Sponsors
ACM: Association for Computing Machinery
SIGARCH: ACM Special Interest Group on Computer Architecture
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 14,   Downloads (12 Months): 73,   Citation Count: 0
Additional Information:

abstract   references   index terms   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/1542275.1542298
What is a DOI?

ABSTRACT

"Is transactional memory useful?" is the question that cannot be answered until we provide substantial applications that can evaluate its capabilities. While existing TM applications can partially answer the above question, and are useful in the sense that they provide a first-order TM experimentation framework, they serve only as a proof of concept and fail to make a conclusive case for wide adoption by the general computing community.

This paper presents QuakeTM, a multiplayer game server; a complex real life TM application that was parallelized from the sequential version with TM-specific considerations in mind. QuakeTM consists of 27,600 lines of code spread across 49 files and exhibits irregular parallelism for which a task parallel model fits well. We provide a coarse-grained TM implementation characterized with eight large transactional blocks as well as a fine-grained implementation which consists of 58 different critical sections and compare these two approaches. In spite of the fact that QuakeTM scales, we show that more effort is needed to decrease the overhead and the abort rate of current software transactional memory systems to achieve a good performance. We give insights into development challenges, suggest techniques to solve them and provide extensive analysis of the transactional behavior of QuakeTM, with an emphasis and discussion of the TM promise of making parallel programming easier.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

1
 
2
A. Abdelkhalek and A. Bilas. Parallelization and performance of interactive multiplayer game servers. In IPDPS'04: Proc. 18thIinternational Parallel and Distributed Processing Symposium, Apr. 2004, pages 72--81.
 
3
A. Abdelkhalek, A. Bilas and A. Moshovos. Behavior and performance of interactive multi-player game servers. In ISPASS'01: Proc. of the 2001 International IEEE Symposium on Performance Analysis of Systems and Software, Nov. 2001, pages 355--366.
4
 
5
 
6
L. Dalessandro, V. J. Marathe, M. F. Spear and M. L. Scott. Capabilities and Limitations of Library-Based Software Transactional Memory in C++. In TRANSACT'07: Proc. 2nd ACM SIGPLAN Workshop on Transactional Computing, Aug 2007.
7
8
9
10
11
 
12
M. McGuire. Quake 2 BSP File Format. http://www.flipcode.com/archives/Quake_2_BSP_File_Format.shtml
 
13
C. C. Minh, J. Chung, C. Kozyrakis and K. Olukotun. STAMP: Stanford transactional applications for multi-processing. In IISWC'08: Proc. 11th IEEE International Symp. on Workload Characterization, Sep. 2008, pp. 35--46.
14
15
16
 
17
 
18
19
20
21

Collaborative Colleagues:
Vladimir Gajinov: colleagues
Ferad Zyulkyarov: colleagues
Osman S. Unsal: colleagues
Adrian Cristal: colleagues
Eduard Ayguade: colleagues
Tim Harris: colleagues
Mateo Valero: colleagues