Inhalt

•  
   Introduction
•  
   Scope of this Work
•  
   Document Structure
•  
  Foundations
•  
   Resource Management
•  
   Grid Computing
•  
   First Generation
•  
   Second Generation
•  
   Third Generation
•  
   Commercial Grids
•  
   Service Level Agreements
•  
   Basics
•  
   Content of an Agreement
•  
   Initiation of Negotiation
•  
   WS-Agreement and WS-AgreementNegotiation Protocol
•  
  Structure of a Service Level Agreement
•  
   Context
•  
   Terms
•  
   Guarantee Terms
•  
   Negotiation of Service Level Agreements
•  
  SLA-aware Scheduling
•  
   Negotiation Requirements of Broker Services
•  
   Levels of Service Quality
•  
   SDTs on Resource Specific QoS
•  
   SDTs on Guarantee Level
•  
   SDTs on System Policies
•  
  Phases of Operation
•  
   Negotiation
•  
   Pre-Runtime Phase
•  
   Runtime Phase
•  
   Post-Runtime Phase
•  
  Timing Aspects of Runtime Phases
•  
   Overhead caused by Initialization
•  
   Overhead caused by Checkpointing
•  
   Overhead caused by Migration
•  
   Overhead caused by Restart
•  
   Determining the Minimum Slot
•  
   Determining the Checkpoint Frequency
•  
   Fault Tolerance with intra-cluster scope
•  
  Basic SLA-aware Scheduling
•  
   Input Parameters
•  
   Output Parameters
•  
   Initialization
•  
   Handling of Running and SLA-Reservation Jobs
•  
   Handling of Problematic Jobs
•  
   Handling of SLA-Requests on Reservations
•  
   Handling of SLA-Non-Reservation Jobs
•  
   Handling of regular SLA-Requests
•  
   Handling of existing and requested Best-Effort Jobs
•  
   SLA-aware Scheduling with Job Suspension
•  
   Suspension of Best-Effort Jobs
•  
   Suspension of SLA-bound Jobs
•  
   Partial Execution
•  
   Increasing Level of Fault Tolerance
•  
   Increasing Utilization
•  
   Buffer Nodes
•  
   Fault Tolerance with cross-border scope
•  
   Identification of Migration Candidates
•  
   Application of Customer Defined Policies
•  
   Generation of Compatibility Profiles
•  
   Static Profile Matching
•  
   Filtering of Static Results
•  
   Negotiation and Migration
•  
   Migration of Problematic Jobs
•  
   Migration of General Jobs
•  
   Design Aspects
•  
   The HPC4U Project
•  
   Resource Management System
•  
   Grid Interface (GI) with Negotiation Manager (NM)
•  
   Planning Manager (PM) and Machine Manager (MM)
•  
   Access Manager (AM)
•  
   Migration Controller (MC)
•  
   Configuration Manager (CM)
•  
   Execution Manager (EM)
•  
   Subsystem Controller (SSC)
•  
   Checkpoint Generation
•  
   Compatibility Profile
•  
  Architecture and System Properties
•  
   Operating System
•  
   Checkpoint System
•  
   Processor Architecture
•  
   Other System Properties
•  
   Libraries
•  
   Results and Perspectives
•  
   Fault Tolerance Provision
•  
  Use-Case Experiences
•  
   Swedish Meteorological and Hydrological Institute
•  
   Fault Tolerance Aspects
•  
  Centre of Excellence in Aeronautical Research
•  
   Fault Tolerance Aspects
•  
   Key Requirement Security
•  
  Related Work
•  
  Service Level Agreements in Grid Middleware
•  
   Advance Reservations and GARA
•  
   Service Negotiation and Allocation Protocol
•  
  Grid Broker Services
•  
   GRUBER/DI-GRUBER SLA broker
•  
   White Rose Grid
•  
   NextGRID
•  
   SLBSH Project
•  
   Grid Economy
•  
   Process Checkpointing
•  
   Berkeley Labs Checkpointing and Restart (BLCR)
•  
   IBM Metacluster
•  
   Conclusion
•  
   Bibliography