![](bg5.png)
A number of efforts similar to StratUm and GJMF
exist and include, e.g., Falkon (Raicu et al., 2007),
a lightweight task execution framework designed for
Many Task Computing (Raicu et al., 2008). Falkon
and StratUm are similar in use of custom protocols
and service interfaces, but while Falkon is designed
for high submission throughput StratUm is more fo-
cused on raising computation abstraction levels.
GridSAM (Lee et al., 2005) is a standards-based
grid job submission system that abstracts underlying
resource managers through a Web Service interface
built on staged event-driven architecture (SEDA). The
job submission pipeline of GridSAM is similar to the
integration of StratUm and GJMF, but the focus of
StratUm and GJMF lie more towards automation and
adaptability in system architecture design.
The Simple API for Grid Applications
(SAGA) (Kaiser et al., 2006) is an API stan-
dardization initiative that like StratUm and the GJMF
aims to provide a unified interface to grid integration.
The philosophy of the SAGA API differs from
StratUm in that StratUm focuses on automation
and minimization of integration complexity. Per-
formance evaluations comparing StratUm to SAGA
implementations are subject for future work.
For reasons of brevity, the list is far from com-
plete. Compared to most approaches, the aim of this
work is directed more towards design patterns for ab-
straction of complexity in computation management,
and exploration of approaches suitable for the specific
use cases of eScience applications. For more exhaus-
tive treatment of grid job management mechanisms
readers are referred to (
¨
Ostberg and Elmroth, 2010).
6 CONCLUSIONS
In this paper we explore an approach to integration
of eScience applications with distributed computing
environments based on separation of computation de-
sign from computation management. Emphasis of the
work is placed on architectures that raise computa-
tion management abstraction levels and facilitate ver-
satility in computation enactment. As part of the re-
sults we present StratUm, an integration architecture
designed to abstract complexity and raise abstraction
levels in distributed computation management.
ACKNOWLEDGEMENTS
The authors thank Mikael
¨
Ohman, Sebastian Gr
¨
ohn,
and Anders H
¨
aggstr
¨
om for work related to the
project. This work is done in collaboration with
the High Performance Computing Center North
(HPC2N) and is funded by the Swedish National In-
frastructure for Computing (SNIC), the Swedish Gov-
ernment’s strategic research project eSSENCE, the
Swedish Royal Academy of Sciences, and U.S NSF
Grant DMS-1001012, U.S. NIH Grant R01EB7511,
U.S. DOE Award DE-FG02-04ER25621, U.S. NSF
IGERT DGE-02-21715, Institute for Collaborative
Biotechnologies Grant DAAD19-03-D-0004 from the
U.S. Army Research Office.
REFERENCES
Bunch, C., Chohan, N., Krintz, C., and Shams, K. (2011).
Neptune: a domain specific language for deploy-
ing hpc software on cloud platforms. In Proceed-
ings of the 2nd international workshop on Scientific
cloud computing, ScienceCloud ’11, pages 59–68,
New York, NY, USA. ACM.
Casanova, H., Berman, F., Bartol, T., Gokcay, E., Se-
jnowski, T., Birnbaum, A., Dongarra, J., Miller, M.,
Ellisman, M., Faerman, M., Obertelli, G., Wolski, R.,
Pomerantz, S., and Stiles, J. (2004). The virtual in-
strument: Support for grid-enabled mcell simulations.
International Journal of High Performance Comput-
ing Applications, 18(1):3–17.
Drawert, B., Engblom, S., and Hellander, A. (2011). UR-
DME 1.1: User’s manual. Technical Report 2011-003,
Department of Information Technology, Division of
Scientific Computing, Uppsala University.
Foster, I. (2005). Globus toolkit version 4: Software for
service-oriented systems. In Jin, H., Reed, D., and
Jiang, W., editors, IFIP International Conference on
Network and Parallel Computing, LNCS 3779, pages
2–13. Springer-Verlag.
Kaiser, H., Merzky, A., Hirmer, S., Allen, G., and Seidel,
E. (2006). The saga c++ reference implementation:
a milestone toward new high-level grid applications.
In Proceedings of the 2006 ACM/IEEE conference on
Supercomputing, SC ’06, New York, NY, USA. ACM.
Lee, W., McGough, A. S., and Darlington, J. (2005). Per-
formance evaluation of the GridSAM job submission
and monitoring system. In UK e-Science All Hands
Meeting, pages 915–922.
¨
Ostberg, P.-O. and Elmroth, E. (submitted, 2010).
GJMF - A Composable Service-Oriented Grid Job
Management Framework. Preprint available at
http://www.cs.umu.se/ds.
Raicu, I., Foster, I., and Zhao, Y. (2008). Many-task com-
puting for grids and supercomputers. In Workshop on
Many-Task Computing on Grids and Supercomputers
(MTAGS) 2008., pages 1–11.
Raicu, I., Zhao, Y., Dumitrescu, C., Foster, I., and Wilde,
M. (2007). Falkon: a Fast and Light-weight tasK ex-
ecutiON framework. In Proceedings of IEEE/ACM
Supercomputing 07.
BIOINFORMATICS 2012 - International Conference on Bioinformatics Models, Methods and Algorithms
294