supported by the EC IST Programme
CoreGRID European Research Network on Foundations, Software Infrastructures and Applications
for large scale distributed, GRID and Peer-to-Peer Technologies
image
Home arrow Institutes arrow Institute on RMS  

spacer spacer
 
Main Menu
Home
News
Events
CoreGRID WG
CoreGRID NoE
Institutes
Integration Activities
Dissemination
Training & Education
CoreGRID & Industry
Mobility Portal
Trust&Security Portal
Collaboration Gateway
Other Collaborations
Links
 
spacer spacer
spacer spacer
 
Institute on Resource Management and Scheduling
Research Objectives
Institute leader: Ramin Yahyapour (ramin.yahyapourImageudo.edu), University of Dortmund
    Image 

The institute on resource management and scheduling aims at delivering a comprehensive model on the resource management and scheduling for next generation Grids. The management and coordination of resources in a Grid environment is a key topic for next generation Grids (NGG), which has also been identified by the NGG expert group. Currently, there is neither a coherent and generally accepted infrastructure to manage and schedule resources nor are there efficient coordination algorithms that suit the complex requirements of a large scale Grid environment with different resource types. An efficient and transparent Grid resource management is essential for any large scale Grid as the availability of a large amount of resources connected by a network does not constitute a Grid. Key issues in this area are, for instance, the consideration of a cost and accounting model as well as the support for individual policies of Grid users and resource providers.

The members of the research institute will jointly address all relevant scheduling and resource management topics to define a common and generic solution for Next Generation Grids. This includes the architectural perspective which has to support scalability and cooperation across different administrative domains. As a Grid is dynamic by nature, it is essential to discover the available resources, coordinate the resources required by a complex job and monitor the progress of this job. This complex and cumbersome process cannot be handled by the user directly. Therefore, a large-scale Grid requires an elaborate resource management system that hides this process from the user. To this end, new scheduling algorithms and scheduling policies will be developed for coordinating the access to resources and for supporting complex job requirements. The actual management of resources and jobs further needs several services that provide the necessary functionality e.g. to manage the scheduling of complex job workflows.

Due to the different sub problems in providing a complete solution for Grid resource management and scheduling, the following key tasks have been identified:

  1. Definition of a Grid scheduling architecture
  2. Provision of higher-level scheduling models and algorithms
  3. Realization of complex coordination of resources including workflow scheduling
  4. Provision of fault-tolerant and reliable scheduling methods
  5. Evaluation of Grid Scheduling Strategies
  6. Definition of a model for coordinated scheduling for all resource types, especially including data and network scheduling
  7. Identification of requirements and strategies for supporting Grid business models within the scheduling and resource management process
  8. Service Level Agreements
  9. Virtualization Management
Roadmap version 3 on Resource Management and Scheduling

Publications related to the Institute on Resource Management and Scheduling
 
Research Groups

Research Groups
Participants
Grid Scheduling Architecture
CNR-ISTI, TUD, FHG, FZJ, UNIDO, UPC, UNICAL, PSNC
Reservation-based Scheduling
CNRS, ZIB, VUA
Examination of Economic-Based Scheduling Mechanisms
CNR-ISTI, FHG, FZJ, UNIDO, PSNC, CNRS
Integration of the eNANOS execution environment with the
GRMS Resource Broker
UPC, PSNC
Integration of the advanced reservations mechanism
between the GRMS and eNANOS Brokers
UPC, PSNC
Negotiation and Interaction of Scheduler in GRMS
UNIDO, PSNC, VUA
Grid Scheduling Ontology
FhG SCAI, FZJ, UoM
Integration of ISS into the VIOLA/MetaScheduling
Environment
HES-SO/EPFL, FhG SCAI, FZJ 
Workflow Scheduling with P-GRADE and KOALA SZTAKI, TUD, UMUE
Review of Workflow Description Languages for Grid
Scheduling
UNIDO, FHG
Automatic mapping of applications CNR-ISTI, CNRS 
Job co-allocation  CNRS, TUD
Scheduling of high performance parallel applications  ISTI-CNR, DI-UNIPI, UNIDO, CNRS
Study of performance models and contracts for a class of
Grid application 
ISTI-CNR, DI-UNIPI, UNIDO, VUA
Collaboration for a Grid Evaluation and Benchmarking
System
TUD, UNIDO 
Standard Grid Workload Format  UOW, UPC, PSNC 
Performance prediction using the Gamma model  EPFL, PSNC, UPC, UOW, WCNS 
Application of the GPRES performance prediction system
to analysis of real LoadLeveler workloads
PSNC, UPC, WCNS 
Handling performance prediction errors in Grid
Scheduling
ISTI-CNR, PSNC, WCNS 
Validation of the ISS Multi-level Scheduling System  CETIC, HES-SO, EPFL, INRIA 
 
Latest Research Highlights

Investigating peer-to-peer meta-brokering in Grids

  CoreGRID Technical Report TR-0170 [pdf]:

Grid Computing has succeeded in establishing production Grids serving various user communities all around the world. The emerging Web technologies have already affected Grid development; the latest solutions from peer-topeer networking also need to be considered in order to successfully transform the currently separated production Grids to a World-Wide Grid solving the big problem of Grid Interoperability. This paper gathers the promising peerto- peer technologies and investigates their applicability in current Grids. We state the requirements of a scalable and adaptable high-level brokering architecture, and dene the necessary routing mechanisms that can efciently operate this peer-to-peer Grid meta-brokering architecture.
Image

ExtendingWS-Agreement for dynamic negotiation of Service Level Agreements

 CoreGRID Technical Report TR-0172 [pdf]:

Services in commercial or scientific environments often need to be delivered at a guaranteed service level. This becomes even more important, if composition of services is required, as results will become available for the requester within a predictable timeframe only if the orchestration process may create Service Level Agreements (SLAs) for the individual services. WS-Agreement is a proposed recommendation of the Open Grid Forum (OGF) defining a language and a protocol to create SLAs. As the current version of the specification of WS-Agreement provides only a simple, one-step approach for the creation of SLAs the GRAAP working group of the OGF started working on a more sophisticated protocol allowing multi-step negotiations to create agreements. In this paper we present the approach for negotiations of SLAs using WS-Agreement under specification at the OGF. We describe the state of the work and give an outlook of the next steps until the specification will become a proposed recommendation.

 Image

Considerations for negotiation and monitoring of Service Level Agreements

  CoreGRID Technical Report TR-0167 [pdf]:

Service Level Agreements (SLA) may be used to establish agreements on the quality of a service between a service provider and a service consumer. This report gives an overview on current work done on negotiation and re-negotiation of SLAs within the CoreGRID Institute on Resource Management and Scheduling and in the Grid Resource Allocation Agreement Protocol Working Group (GRAAP-WG) of the Open Grid Forum (OGF). An introduction is given on the proposed OGF recommendation WS-Agreement and its state, followed by a presentation of use-cases for negotiation stemming from from implementations of WS-Agreement. We then describe the approaches currently under discussion in the GRAAP-WG to be consolidated to a negotiation protocol for Service Level Agreements using WS-Agreement. Since the working group of the OGF has not yet consolidated the approaches presented there is not yet a single proposal for a negotiation protocol. However, during the last meeting of the working group the members identified the three approaches that will now be used as a base to specify the protocol for negotiation and re-negotiation of SLAs. Furthermore, the report discusses built-in functionality ofWS-Agreement and additional requirements for the monitoring of SLAs.
Image
 
 
spacer spacer
spacer spacer
 
© 2012 CoreGRID Network of Excellence - European Grid Research
Joomla! is Free Software released under the GNU/GPL License.
 
spacer spacer