Whodunit: transactional profiling for multi-tier applications

ABSTRACT

This paper is concerned with performance debugging of multi-tier applications, such as commonly found in servers and dynamic-content web sites. Existing tools and techniques for profiling such applications are not general enough to track and profile transactions in a generic multi-tier application. We propose transactional profiling that provides a general solution to this problem. We provide novel algorithms and techniques to track and profile transactions that flow through shared memory, events, stages or via interprocess communication using messages. We also measure interference among concurrent transactions.

We describe the design and implementation of Whodunit, our prototype transactional profiler. We demonstrate the correctness of our proposed algorithm for tracking transaction flow through shared memory using Apache and MySQL. Using Whodunit we are able to track and profile transactions that flow through shared memory, events, stages or via message passing, and measure the interference among concurrent transactions. We illustrate the use of Whodunit in obtaining the transactional profile of web servers, a web proxy cache and a bookstore application.

REFERENCES

• 1: ONLamp.com. At http://www.onlamp.com.
• 2: Squid Web Proxy Cache. At http://www.squid-cache.org.
• 3: TPC-W Code Release. At http://www.cs.rice.edu/CS/Systems/DynaServer/TPC-W/.
• 4: Marcos K. Aguilera , Jeffrey C. Mogul , Janet L. Wiener , Patrick Reynolds , Athicha Muthitacharoen, Performance debugging for distributed systems of black boxes, Proceedings of the nineteenth ACM symposium on Operating systems principles, October 19-22, 2003, Bolton Landing, NY, USA
• 5: Glenn Ammons , Thomas Ball , James R. Larus, Exploiting hardware performance counters with flow and context sensitive profiling, Proceedings of the ACM SIGPLAN 1997 conference on Programming language design and implementation, p.85-96, June 16-18, 1997, Las Vegas, Nevada, United States
• 6: Paul Barham , Austin Donnelly , Rebecca Isaacs , Richard Mortier, Using magpie for request extraction and workload modelling, Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation, p.18-18, December 06-08, 2004, San Francisco, CA
• 7: Bellard, F. QEMU open source processor emulator. http://fabrice.bellard.free.fr/qemu/.
• 8: Bernat, A. R., and Miller, B. P. Incremental call-path profiling. Tech. rep., University of Wisconsin, February 2004.
• 9: Corporation, B. S. ServerTrace. http://www.borland.com/us/products/servertrace/index.html.
• 10: Ryan Cunningham , Eddie Kohler, Making events less slippery with eel, Proceedings of the 10th conference on Hot Topics in Operating Systems, p.3-3, June 12-15, 2005, Santa Fe, NM
• 11: Luiz DeRose , Felix Wolf, CATCH - A Call-Graph Based Automatic Tool for Capture of Hardware Performance Metrics for MPI and OpenMP Applications, Proceedings of the 8th International Euro-Par Conference on Parallel Processing, p.167-176, August 27-30, 2002
• 12: Nathan Froyd , John Mellor-Crummey , Rob Fowler, Low-overhead call path profiling of unmodified, optimized code, Proceedings of the 19th annual international conference on Supercomputing, June 20-22, 2005, Cambridge, Massachusetts [doi>10.1145/1088149.1088161]
• 13: Susan L. Graham , Peter B. Kessler , Marshall K. Mckusick, Gprof: A call graph execution profiler, Proceedings of the 1982 SIGPLAN symposium on Compiler construction, p.120-126, June 23-25, 1982, Boston, Massachusetts, United States
• 14: Robert J. Hall, Call path profiling, Proceedings of the 14th international conference on Software engineering, p.296-306, May 11-15, 1992, Melbourne, Australia [doi>10.1145/143062.143147]
• 15: Hall, R. J., and Goldberg, A. J. Call path profiling of monotonic program resources in UNIX. In Proceedings of the USENIX Summer Technical Conference (1993).
• 16: Minwen Ji , Edward W. Felten , Kai Li, Performance measurements for multithreaded programs, Proceedings of the 1998 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems, p.161-170, June 22-26, 1998, Madison, Wisconsin, United States
• 17: Allen D. Malony , Sameer Shende , Robert Bell , Kai Li , Li Li , Nick Trebon, Advances in the TAU performance system, Performance analysis and grid computing, Kluwer Academic Publishers, Norwell, MA, 2004
• 18: Performant, I. Optibench. http://www.performant.com.
• 19: Provos, N. Libevent - an event notification library. Version 1.1 is available from the author's web site, http://www.monkey.org/~provos/libevent/, Oct. 2003. Libevent is also included in recent releases of the NetBSD and OpenBSD operating systems.
• 20: Quest Software, I. PerformaSure J2EE Application Performance Diagnostics, http://www.quest.com/performasure.
• 21: Patrick Reynolds , Janet L. Wiener , Jeffrey C. Mogul , Marcos K. Aguilera , Amin Vahdat, WAP5: black-box performance debugging for wide-area systems, Proceedings of the 15th international conference on World Wide Web, May 23-26, 2006, Edinburgh, Scotland [doi>10.1145/1135777.1135830]
• 22: Sun Developer Network. Java Enterprise Edition. http://Java.sun.com/javaee/.
• 23: (TPC), T. P. P. C. TPC BENCHMARK W (web commerce). At http://www.tpc.org/tpcw/, Feb. 2002.
• 24: Jeffrey S. Vetter , Michael O. McCracken, Statistical scalability analysis of communication operations in distributed applications, Proceedings of the eighth ACM SIGPLAN symposium on Principles and practices of parallel programming, p.123-132, June 2001, Snowbird, Utah, United States
• 25: Matt Welsh , David Culler , Eric Brewer, SEDA: an architecture for well-conditioned, scalable internet services, Proceedings of the eighteenth ACM symposium on Operating systems principles, October 21-24, 2001, Banff, Alberta, Canada
• 26: WWW.FREEBSD.ORG. sys/queue.h Man Page. http://www.freebsd.org/cgi/man.cgi?query=queue&apropos=O&sektion=3&manp%ath=FreeBSD+6. 1-RELEASE&format=html.

INDEX TERMS

Primary Classification:
  D. Software
  D.2 SOFTWARE ENGINEERING
      D.2.8 Metrics
          Subjects: Performance measures

General Terms:
Algorithms, Design, Performance

Keywords:
distribution, profiling