K-WfGrid
K-WfGrid is an EU IST STREP project (*IST*-2002-511385).
The goal of this project
is to deliver a robust Grid environment which provides a user friendly
interface for the composition of the Grid workflow applications, performs smart
scheduling of application components and finally executes them on the available Grid resources.
All phases of an application processing will be strongly supported by the domain specific
knowledge and semantic technologies.
The target applications could be from different scientific and business domains.
For instance, scientific simulations like flood forecasting simulation and
industrial applications like ERP and traffic management.
Besides taking part in project management, coordination and
dissemination tasks, DPS group is also involved in the two technical work
packages (WPs) of K-WfGrid, that is WP2 and WP3. These work packages
target development of a scheduler and performance monitoring and analysis service, respectively.
Fig. 1 depicts the workflows of the Grid services (see below).

Fig.1 K-Wf Grid system architecture.
|

Fig.2 Scheduler in K-Wf Grid system.
|
WP2: Workflow Orchestration and Execution Environment
Within WP2, DPS is developing a performance-driven scheduler.
Scheduler is a service responsible for actual mapping of
the workflow activities onto the Grid resources. It determines which
instances of a set of alternative Web Service Operation (WSO) instances
will be selected for invocation in the current workflow execution.
Scheduling decisions are supported by the knowledge collected in the general purpose
knowledge-base, called GOM, and by the Performance Analysis and Monitoring service.
WP3: Performance Monitoring and Analysis for Grid Infrastructures and Applications
DPS is performing leading role in the work package 3. That means, along with research and
development, DPS is coordinating the whole work conducted in WP3, as well.
DPS also focuses on studying important performance and
dependability metrics along with their description, measurement and association with the
Grid workflows.
The performance of workflows will be
examined through workflow-specific metrics. We introduce a novel ontology for
performance data associated with the Grid workflows. This
ontology describes well-defined concepts associated with Grid workflows
and performance metrics associated with these concepts.
Our main aim is to develop a set of distributed performance analysis
services (DIPAS). These services collect performance and monitoring
data from the monitoring service and provide performance results to the
client of the performance analysis service during the runtime of the
workflows. DPS group aims at supporting performance interpretation and
bottleneck search. It will develop a specification language for
performance properties in combination with search processes for
performance problems. We will also support interpretation of
performance data.
DPS works on data representations and service interfaces.
Performance, monitoring and event data can be represented in a
well-defined XML format, thus interaction with other services is
substantially simplified and becomes consistent. Instrumentation and measurement requests are
issued dynamically via an XML interface.
Moreover, DPS also conducts research in developing adaptive
monitoring for the Grid and instrumentation techniques for multilingual
applications. This work is done with CYFRONET group in developing the
Grid Monitoring and Instrumentation Service.
Posters and Demonstration movies
Posters
Demonstration movies
- Flood applications
- Coordinated Traffic Management (CTM)
- Enterprise Resource Planning (ERP)
- Modelling, Monitoring and Analysis of Dependability Metrics (QoSMAT)
- Short demo:
1420x980 (AVI, 38MB),
1420x980 (SWF, 6MB),
1024x768 (AVI, 60MB),
1024x768 (SWF, 8MB)
- Generating a dependability graph:
1420x980 (AVI, 20MB),
1420x980 (SWF, 3MB)
- Adding additional dependencies:
1420x980 (AVI, 28MB),
1420x980 (SWF, 8MB)
- Status update and propagation:
1420x980 (AVI, 28MB),
1420x980 (SWF, 8MB)
- Metric-based view:
1420x980 (AVI, 65MB),
1420x980 (SWF, 14MB)
- Status modelling:
1420x980 (AVI, 18MB),
1420x980 (SWF, 2MB)
Software download
Use this settings for downloading software.
People involved
Past members:
For further information about DPS work on K-WfGrid please contact: Thomas Fahringer or Hong-Linh Truong.
Scientific contributions of DPS to K-Wfgrid
- Hong-Linh Truong, Schahram Dustdar, Thomas Fahringer "Performance Metrics and Ontology for Grid Workflows", (Submitted PDF), Future Generation of Computer Systems, (c) Elsevier, 2007. Accepted.
- Robert Samborski, Hong-Linh Truong, Thomas Fahringer,Monitoring and Analysis of Dependability Metrics of
Grid Services (PDF Poster), Cracow Grid workshop, October 16-18, 2006, Cracow, Poland
- Francesco Nerieri, Radu Prodan, Thomas Fahringer,Kalipy: A Tool for Online Performance Analysis of
Grid Workflows through Event Correlation(Submitted PDF),2nd
IEEE International Conference on e-Science and Grid Computing, (c) IEEE
Computer Society Press, Dec. 4- 6, 2006, Amsterdam, Netherlands.
- Hong-Linh Truong, Robert Samborski, Thomas Fahringer, Towards a Framework for Monitoring and Analyzing QoS Metrics of Grid
Services
(Submitted PDF), 2nd IEEE International Conference on e-Science and Grid Computing, (c) IEEE Computer Society Press, Dec. 4- 6, 2006, Amsterdam, Netherlands.
- Hong-Linh Truong, Peter Brunner, Thomas Fahringer, Francesco Nerieri, Robert Samborski, Bartosz Balis,
Marian Bubak, Kuba Rozkwitalski, K-WfGrid Distributed Monitoring and Performance Analysis Services for Workflows in the Grid,
(Submitted PDF), 2nd IEEE International Conference on e-Science and Grid Computing, (c) IEEE Computer Society Press, Dec. 4- 6, 2006, Amsterdam, Netherlands.
- Farrukh Nadeem, Muhammad Murtaza Yousaf, Radu Prodan, Thomas Fahringer, Soft Benchmarks-based Application Performance Prediction using a Minimum Training Set, (Submitted PDF),
2nd IEEE
International Conference on e-Science and Grid Computing (e-Science
2006), (C) IEEE Computer Society Press, December 2006, Amsterdam,
The Netherlands.
- Michael Mair, Jun Qin, Marek Wieczorek and Thomas Fahringer.
Workflow Conversion and Processing in the ASKALON Grid Environment, 2nd
Austrian Grid Symposium, September 2006, Innsbruck, Austria
- Marek Wieczorek, Mumtaz Siddiqui, Alex Villazon, Radu Prodan, Thomas Fahringer, Applying Advance Reservation to
Increase Predictability of Workflow Execution on the Grid,(Submitted PDF),
2nd IEEE
International Conference on e-Science and Grid Computing (e-Science
2006), (C) IEEE Computer Society Press, December 2006, Amsterdam,
The Netherlands.
- Dieter Theiner, Marek Wieczorek, Reduction of Calibration Time of Distributed Hydrological Models by Use of Grid Computing and Nonlinear Optimisation Algorithms, 7th International Conference on Hydroinformatics (HIC 2006), September 2006, Nice, France
- Marek Wieczorek, Radu Prodan, Thomas Fahringer, Dynamic Programming Based Approach for Bicriteria Workflow Scheduling on the Grid, Poster on HPDC 2006.
- Bartosz Balis, Hong-Linh Truong, Marian Bubak, Thomas Fahringer,Krzysztof Guzy, Kuba Rozkwitalski,
"An Instrumentation Infrastructure for Grid Workflow Applications", (Submitted PDF),
International Symposium on Grid Computing, High-Performance and
Distributed Applications (GADA06), LNCS, (c) Springer-Verlag, November
2-3, 2006, Montpellier, France.
- Peter Brunner, Hong-Linh Truong, Thomas Fahringer "Performance Monitoring and Visualization of Grid
Scientific Workflows in ASKALON", (Submitted PDF),
The 2nd International Conference on High Performance Computing and
Communications (HPCC-06), LNCS, (c) Springer-Verlag, Munich, 13th-15th
September, 2006.
- Francesco Nerieri, Radu Prodan, Thomas Fahringer, Hong-Linh Truong, "
Performance Analysis of Grid Workflow Applications", (Submitted PDF),7th
IEEE/ACM International Conference on Grid Computing (Grid'06), (C) IEEE
Computer Society Press, September 28th-29th, 2006, Barcelona, Spain.
-
Rubing Duan, Radu Prodan, Thomas Fahringer, Run-time Optimization for Grid Workflow Applications (Submitted PDF),
7th IEEE/ACM International Conference on Grid Computing (Grid'06), (C)
IEEE Computer Society Press, September 2006, Barcelona, Spain.
- Rubing Duan, Radu Prodan, Thomas Fahringer, Data Mining-based Fault Prediction and Detection on the Grid (Submitted PDF),
15th IEEE International Symposium on High Performance Distributed
Computing (HPDC'06), (C) IEEE Computer Society Press, June 2006, Paris,
France.
- Thomas Fahringer, Radu Prodan, Juergen Hofer, Farrukh Nadeem,
Francesco Nerieri, Stefan Podlipnig, Jun Qin, Duan Rubing, Mumtaz
Siddiqui, Hong-Linh Truong, Alex Villazon, Marek Wieczorek, ASKALON: A Development and Grid Computing Environment for Scientific Workflows, In: Workflows for eScience, XXX (Eds.), Springer, 2006 (accepted).
- B. Balis, M. Bubak, J. Dziwisz, H.-L. Truong, and T. Fahringer, "
Integrated Monitoring Framework for Grid Infrastructure and Applications", (
PDF), In P. Cunningham and M. Cunnin
gham, editors, Innovation and the Knowledge Economy. Issues, Applications, Ca
se Studies, pages 269-276, Ljubljana, Slovenia, October 2005. IOS Press.
- Marek Wieczorek, Radu Prodan, Thomas Fahringer, "Scheduling of Scientific Workflows in the ASKALON Grid Environment", ACM SIGMOD Record, 09/2005.
- Hong-Linh Truong, Bartosz Balis, Marian Bubak, Jakub Dziwisz,
Thomas Fahringer, Andreas Hoheisel, "Towards Distributed Monitoring and Perfo
rmance Analysis Services in the K-WfGrid
Project", (PDF, PS.GZ), In Proceedings of 6th
International Conference on Parallel Processing and Applied Mathematics (PPAM'20
05), LNCS, (c) Springer-Verlag, Poznan, Poland, 11-14 September, 2005.
-
Marek Wieczorek, Radu Prodan, Thomas Fahringer, "Comparison of Workflow Scheduling Strategies on the Grid",
In Proceedings of 6th International Conference on Parallel Processing
and Applied Mathematics (PPAM 2005), LNCS 3911, ) Springer-Verlag.
- Hong-Linh Truong, Thomas Fahringer, Schahram Dustdar, "Dynamic Instrumentation,
Performance Monitoring and Analysis of Grid Scientific Workflows", (PDF), 3(1-2):1-18, Journal of Grid Computing, (c) Springer Netherlands, June 2005.
- Hong-Linh Truong, Thomas Fahringer, "Soft Computing Approach to Performance Analysis of Parallel and Distributed Programs", (PDF), 11th International Euro-Par Conference (Euro-Par 2005), LNCS, (c) Springer-Verlag, Lisboa, Portugal, 30th August- 2nd September 2005.
- Hong-Linh Truong, Thomas Fahringer, Francesco Nerieri, Schahram Dustdar,"Performance Metrics and Ontology for Describing Performance Data of Grid
Workflows", (PDF), IEEE International Symposium on Cluster Computing
and Grid 2005 (CCGrid2005),1st
International Workshop on Grid Performability, IEEE
Computer Society Press, Cardiff, UK, 9 - 12 May 2005.
- Hong-Linh Truong, Thomas Fahringer, "Online Performance Monitoring and Analysis of Grid Scientific Workflows", (PDF), European Grid Conference 2005 (EGC2005), LNCS,(c) Springer-Verlag, Amsterdam,
The Netherlands, February 14 -16, 2005.
Deliverables
- Hong-Linh Truong, Thomas Fahringer and Bartosz Balis, Performance Monitoring and Analysis Services - Stable Software, K-WfGrid D3 Deliverable, August, 2006.
- Hong-Linh Truong, Peter Brunner, Vlad Nae, Robert Samborski, Performance Analysis Service - Developer Manual, K-WfGrid Deliverable, October, 2006.
- Hong-Linh Truong, Robert Samborski, Peter Brunner Performance Analysis Service - User Manual, K-WfGrid Deliverable, February, 2007.
- Hong-Linh Truong, Performance Service Interfaces and Data Representation - Developer Manual, K-WfGrid Deliverable, September, 2006.
- Hong-Linh Truong, Performance Service Interfaces and Data Representation - User Manual, K-WfGrid Deliverable, September, 2006.
- Marek Wieczorek, Scheduler - User Manual, K-WfGrid Deliverable, September, 2006.
- Marek Wieczorek, Scheduler - Developer Manual, K-WfGrid Deliverable, September, 2006.
Last modified: by