The use of the Grid for interactive applications imposes new requirements on tools
supporting performance analysis, which are not well addressed by tools available in
the area of parallel and distributed programming. The two most important demands are
an on-line mode of operation, and the ability to measure user-defined,
application-specific metrics, such as, for example, the response time of a specific
user interaction. The paper starts with an analysis of these requirements, which is
based on real-world applications developed in the European CrossGrid project. The
most important feature of these applications is the presence of a person in the main
computing loop, who inspects intermediate results and controls the further behavior
of the application. Based on the results of the requirements analysis, a new
performance analysis tool G-PM is currently being implemented. The paper outlines
the main ideas as well as some design details of GPM, with a focus on the concept of
user-defined metrics, the possible problems of this concept, and some of its
implementation aspects.
Balaton, Z.
,
Kacsuk, P.
,
Podhorszki, N.
, and
Vajda, F.2000. Comparison of representative Grid monitoring tools.
Technical Report LPDS-2/2000, Laboratory of Parallel and Distributed Systems
(SZTAKI) (ftp://ftp.lpds.sztaki.hu/pub/lpds/publications/reports/lpds-2-2000.pdf).
2.
Balaton, Z.
,
Kacsuk, P.
,
Podhorszki, N.
, and
Vajda, F.
2001.
From cluster monitoring to Grid monitoring based on GRM
. In
R. Sakellariou
,
J. Keane
,
J. Gurd
, and
L. Freeman
, editors, Euro-Par 2001 Parallel Processing, 7th International
Euro-Par Conference, August, Manchester,
UK, Vol. 2150 of Lecture Notes in Computer Science,
Springer-Verlag, Berlin, pp.
874-881
.
3.
Balis, B.
,
Bubak, M.
,
Funika, W.
,
Szepienic, T.
, and
Wismüller, R.
2002.
An infrastructure for Grid application monitoring
. In
D. Kranzlmüller
,
P. Kacsuk
,
J. Dongarra
, and
J. Volkert
, editors, Recent Advances in Parallel Virtual Machine and Message
Passing Interface, 9th European PVM/MPI Users’ Group Meeting,
September, Linz, Austria, Vol. 2474 of
Lecture Notes in Computer Science, Springer-Verlag,
Berlin, pp.
41-49
.
4.
Bubak, M.
,
Funika, W.
,
Iskra, K.
,
Maruszewski, R.
, and
Wismüller, R.
1999.
Enhancing the functionality of performance measurement tools for message
passing environments
. In
J. Dongarra
,
E. Luque
, and
T. Margalef
, editors, Recent Advances in Parallel Virtual Machine and Message
Passing Interface, Proceedings of the 6th European PVM/MPI Users Group
Meeting, September, Barcelona,
Spain, Vol. 1697 of Lecture Notes in Computer Science,
Springer-Verlag, Berlin, pp.
67-74
(http://wwwbode.in.tum.de/%7Ewismuell/pub/europvm99.html).
5.
Bubak, M.
,
Funika, W.
,
Balis, B.
, and
Wismüller, R.2001. Online OCM-based tool support for parallel
applications. In
Y. C. Kwong
, editor, Annual Review of Scalable Computing, Vol. 3, chapter 2,
World Scientific, Singapore
, pp. 32-62.
6.
Bubak, M.
,
Funika, W.
, and
Wismüller, R.
2002.
The CrossGrid performance analysis tool for interactive grid applications
. In
D. Kranzlmller
,
P. Kacsuk
,
J. Dongarra
, and
J. Volkert
, editors, Recent Advances in Parallel Virtual Machine and Message
Passing Interface, 9th European PVM/MPI Users’ Group Meeting,
September, Linz, Austria, Vol. 2474 of
Lecture Notes in Computer Science, Springer-Verlag,
Berlin, pp.
50-60
.
7.
Cargille, J.
and
Miller, B.P.1992.
Binary wrapping: a technique for instrumenting object code
. ACM Sigplan Notices, 27(6):
17-18
.
8.
Fahringer, T.
,
Gerndt, M.
,
Riley, G.
, and
Träff, J.L.1999. Knowledge specification for automatic performance
analysis. APART Technical Report, ESPRIT IV Working Group on Automatic
Performance Analysis (http://www.fz-juelich.de/apart-1/reports/wp2-asl.ps.gz).
9.
Fahringer, T.
and
Seragiotto, C.
2001.
Modeling and detecting performance problems for distributed and parallel
programs with JavaPSL
. In Proceedings of the 9th IEEE High Performance Networking and
Computing Conference, SC’2001, November,
Denver, CO.
10.
Gunter, D.et al.
2000.
NetLogger: a toolkit for distributed system performance analysis
. In Proceedings of the IEEE Mascots 2000 Conference,
August, Lawrence Berkeley National Laboratory Report LBNL-46269.
11.
Hollingsworth, J.R.
,
Miller, B.P.
,
Gonçalves, M.J.R.
,
Xu, Z.
,
Naim, O.
, and
Zheng, L.
1997.
MDL: a language and compiler for dynamic program instrumentation
. In Proceedings of the International Conference on Parallel
Architectures and Compilation Techniques,
November, San Francisco, CA (ftp://grilled.cs.wisc.edu/technical_papers/mdl.ps.gz).
12.
Ludwig, T.
,
Wismüller, R.
,
Sunderam, V.
, and
Bode, A.1997. OMIS - On-line Monitoring Interface Specification (Version
2.0), Vol. 9 of LRR-TUM Research Report Series,
Shaker-Verlag, Aachen
(http://wwwbode.in.tum.de/%7Eomis/OMIS/Version-2.0/version-2.0.ps.gz).
13.
Malony, A.
and
Shende, S.
2000.
Performance technology for complex parallel and distributed systems
. In
G. Kotsis
and
P. Kacsuk
), editors, Proceedings of the 3rd Austrian - Hungarian Workshop on
Distributed and Parallel Systems, DAPSYS 2000, Kluwer,
Dordrecht, pp.
37-46
(http://www.cs.uoregon.edu/research/paracomp/papers/dapsys2k.ps.gz).
Podhorszki, N.
and
Kacsuk, P.
2000.
Design and implementation of a distributed monitor for semi-on-line
monitoring of VisualMP applications
. In Proceedings of DAPSYS’2000 Distributed and Parallel
Systems, From Instruction Parallelism to Cluster Computing,
Balatonfüred, Hungary, pp.
23-32
.
16.
Schwarz, R.
and
Mattern, F.1994.
Detecting causal relationships in distributed computations: in search of
the Holy Grail
. Distributed Computing, 7(3):
149-174
(http://www.inf.ethz.ch/department/IS/vs/publ/papers/holygrail.ps).
17.
Shende, S.
,
Malony, A.D.
,
Cuny, J.
,
Lindlan, K.
,
Beckman, P.
, and
Karmesin, S.
1998.
Portable profiling and tracing for parallel scientific applications using C++
. In Proceedings of the ACM SIGMETRICS Symposium on Parallel and
Distributed Tools (SPDT’98), August,
Welches, OR, pp.
34-145
.
18.
Truong, H.-L.
and
Fahringer, T.
2002.
SCALEA: a performance analysis tool for distributed and parallel programs
. In
B. Monien
and
R. Feldman
, editors, Euro-Par 2002 Parallel Processing, 8th International
Euro-Par Conference, August, Paderborn,
Germany, Vol. 2400 of Lecture Notes in Computer Science,
Springer-Verlag, Berlin, pp.
75-85
.
19.
Vetter, J.
and
Reed, D.2000.
Real-time monitoring, adaptive control and interactive steering of
computational grids
. International Journal of High Performance Computing
Applications, 14:
357-366
.
20.
Wismüller, R.
,
Trinitis, J.
, and
Ludwig, T.
1998.
OCM - a monitoring system for interoperable tools
. In Proceedings of the 2nd SIGMETRICS Symposium on Parallel and
Distributed Tools SPDT’98, August,
Welches, OR, pp.
1-9
(http://wwwbode.in.tum.de/%7Ewismuell/pub/spdt98.ps.gz).
21.
Wolf, F.
and
Mohr, B.
1999.
EARL - a programmable and extensible toolkit for analyzing event traces of
message passing programs
. In
A. Hoekstra
and
B. Hertzberger
, editors, Proceedings of the 7th International Conference on High
Performance Computing and Networking (HPCN 99), Amsterdam, the
Netherlands, pp.
503-512
.
22.
Wolf, F.
and
Mohr, B.
2000.
Automatic performance analysis of MPI applications based on event traces
. In
A. Bode
,
T. Ludwig
,
W. Karl
, and
R. Wismüller
, editors, Euro-Par 2000 Parallel Processing, 6th International
Euro-Par Conference, August, Munich,
Germany, Vol. 1900 of Lecture Notes in Computer
Science, Springer-Verlag, Berlin, pp.
123-132
.
23.
Wolski, R.
,
Spring, N.
, and
Hayes, J.1999.
The Network Weather Service: a distributed resource performance forecasting
service for metacomputing
. Future Generation Computer Systems, 15:
757-768
.
