Research activities
I am currently working as senior researcher at the Hasso-Plattner-Institute. Before that, I was a PostDoc at the Humboldt-University in Berlin, in the Computer Architecture and Communication group. I also work regularly as guest lecturer at the Blekinge Tekniska Högskola in Sweden.
From 2002 – 2007, my primary research focus was in dynamic resource usage for middleware-based distributed systems. From 2005 to 2007, I was a work component leader in the European Adaptive Services Grid (ASG) research project, creating a system model and implementations for grid-enabled SOA infrastructure.
Meanwhile, my focus shifted towards dependability and scalability aspects in modern multi-core architectures. This includes the research on fault prediction and fault injection approaches in modern parallel commodity systems.
Membership
- Co-chair of the DRMAA Working Group at the Open Grid Forum (OGF) standardization body (since 4/2006)
- IEEE, GI
Program Committee / On-Behalf Reviews
- International on Service-Oriented Computing (ICSOC 2010)
- International Symposium on Reliable Distributed Systems (SRDS 09)
- International Conference on Software Engineering and Knowledge Engineering (SEKE 2008, SEKE 2009, SEKE 2010)
- 2nd International Workshop on e-learning and Virtual and Remote Laboratories 2008
- Net.ObjectDays 2006.
- GSEM 2006 and DLS 2006
- GSEM 2005 and Net.ObjectDays 2005.
Earlier Projects
Industry
I organized collaboration projects with Software AG, Intel, Microsoft, Siemens and Daimler Chrysler Research in the past.
Grid-Occam
My Grid-Occam project tried to re-use the old-fashioned transputer programming language Occam to today’s grid environments. It would be great to bring this effort back to life, since we stopped with a reasonably well working Grid-Occam compiler (Java back-end) and a first MPI runtime library. Read the papers for more information, or watch a demo.
Condor
For several years, I was the maintainer for one of the available Condor Debian packages. Condor is a traditional cluster system for high throughput computing in local and wide-area distributed environments. Tim took the responsibility, since I had no longer a need for it after leaving HPI.
BB-Grid
BB-Grid was a initiative from 2005 to 2007, which connected computational resources of HPI, TU-Berlin, BTU Cottbus and the Universitz of Potsdam. We ended up in a working Globus 3 infrastructure, connecting all the clusters.
Adaptive eXecution Platform
From 2004-2007, I was the lead of the Services Infrastructure work component in the Adaptive Services Grid EU project. The project developed an open development platform for the automated and adaptive creation of complex service workflows, based on semantic service descriptions. Our group at HPI provided the Services Infrastructure (SI) component. We continuing the research on this service infrastructure under the new label Adaptive eXecution Platform (AXP).
AXP provides an uniform, WSRF-based access to stateful service instances on dynamically allocated execution resources. This includes functionalities for deployment, instantiation, invocation, and monitoring of internal or proxy services. Our current implementation is based on a combination of commercial-of-the-shelf (COTS) middleware products with established standards from both the Web and the grid service community. It supports dynamic resource allocation and service placement for heterogeneous component technologies like .NET and J2EE.
The underlying system model and the implementations acted as central material for my doctoral thesis. Since the thesis was written in german (yes, in german language), it might be easier to get all relevant details from the according papers and the Semantic Service Provisioning book.
This page has the following sub pages.