Service Availability: 5th International Service Availability Symposium, ISAS 2008 Tokyo, Japan, May 19-21, 2008 Proceedings (Lecture Notes in Computer Science, 5017)

This book constitutes the refereed proceedings of the 5th International Service Availability Symposium, ISAS 2008, held in Tokyo, Japan, in May 2008. The 12 revised full papers presented together with 2 keynote papers and 2 tutorials were carefully reviewed and selected from 28 submissions. The papers are organized in topical sections on enterprise system dependability, software service availability, service availability platform, and service dependability analysis. Title Page Preface Organization Table of Contents The Italian $e$-Government Enterprise Architecture: A Comprehensive Introduction with Focus on the SLA Issue Introduction Strategic and Governance Actions Overview of SPCoop Enteprise Architecture Service Agreements Cooperation Domains and Cooperation Agreements Repositories for Agreements and Schemas/Ontologies Service Level Agreements: A Real Challenge Discussion and Future Work Challenges and Opportunities for Computer Science in Services Science Introduction Emergence of Services Science Opportunities Model-Driven Approach to Services IT Architecture Supporting Services Analytics and Optimization Challenges Service Dependability Understanding and Modeling Human Minds Concluding Remarks – The Changing Roles of Computer Science References Predictive Algorithms and Technologies for Availability Enhancement Achieving and Assuring High Availability Overview Quantified Availability Assurance Recovery from Failures Caused by Mandelbugs Proactive Recovery and Aging-Related Bugs References Optimizing Security Measures in an Intrusion Tolerant Database System Introduction Intrusion Tolerant Database System Basic Concept Model Description Semi-markov Analysis Security Measures System Availability System Integrity Rewarding Availability Numerical Illustrations Parameter Set System Integrity Rewarding Availability Conclusion The Impact of Unavailability on the Effectiveness of Enterprise Information Security Technologies Introduction USB Access Control Solutions Digital Rights Management (DRM) Solutions Disk Encryption Solutions Operating System Solutions Discussion of Technologies Unavailability Impact Analysis Trade-Offs Productivity Loss Departing Users Conclusion Interaction Faults Caused by Third-Party External Systems — A Case Study and Challenges Introduction Scenario Description and a Case Study System Model Errors Caused by Interaction Faults A Case Study Challenges for Error Detection Implicit Redundancies Omissions and Timing Errors Sequence Errors Message Formatting Errors Missing Data Data Syntax Errors Inconsistent Values Action Errors Conclusions and Future Work User-Perceived Software Service Availability Modeling with Reliability Growth Introduction Model Description Model Analysis Distribution of Transition Time of $X(t)$ between State $W$ State Occupancy Probability Software Service Availability Measures Numerical Examples Concluding Remarks Execution Path Profiling for OS Device Drivers: Viability and Methodology Introduction Paper Emphasis and Contributions Paper Organization Related Work System Model: The Entailed OS Kernel Components Developing the Basis for Code Tracing: The I/O Request Packet (IRP) Interface The Processing of I/O Requests Mode, Transition and Operational Profile of a Device Driver Developing the Basis for Code Tracing: The Functional Interface The PE/COFF Executable Format and DLL-Proxying Call Strings as Code Path Abstractions Identifying Execution Hotspots: Call String Clustering Aspects Metrics to Express Call String Similarity Cluster Linkage Methods and Agglomeration Coefficient Evaluating the Viability of the Execution Profiling Methodology Revealing the Execution Hotspots: MDS Plots of the CSs Similarity Cutoffs: Testing Overhead Versus Diversity Masking Discussion and Results Interpretation Conclusions and Future Research Directions Analysis of a Software System with Rejuvenation, Restoration and Checkpointing Introduction Software Availability Modeling Model Description Formulation of System Availability Optimization Algorithm Numerical Examples Exponential System Failure Time Weibull System Failure Time Conclusions A Platform for Cooperative Server Backups Based on Virtual Machines Introduction Cooperative Platform for Service Backups System Architecture Reliable Replication of Virtual Machines Service Placement Membership Experiments Experimental Setup Replication Overheads Virtual Machine Overheads Handling Changes Related Work Conclusions and Future Work Platform Management with SA Forum and Its Role to Achieve High Availability Introduction System Model Information Model for Physical Resources Information Model for Execution Environments Dependencies PLM Objects within the Overall AIS Information Model Operation Administration and Maintenance State Model States for Physical Resources States for Execution Environments Summary State Virtualization Support Health Checking and Fault Analysis Faults of Physical Resources Faults of Execution Environments Correlation Upgrade Support Operating System Upgrade Hardware Upgrade Achieving 5 Nines in Reality Reduce the Time to Repair Avoid Blaming Hardware as Faulty When It Is Not Faulty Summary and Outlook References Automatic Generation of AMF Compliant Configurations Introduction AMF Compliant Configurations: Background and Related Work Background Related Work Configuration Generation Input Data and Validation Type Selection Generating the Remaining AMF Entities Completing the Configuration Attributes The Configuration Generator Tool Description of the Tool The Tool User Interface Application Example Conclusions References Dependability Evaluation of a Replication Service for Mobile Applications in Dynamic Ad-Hoc Networks Introduction Background Scenario Description Background on Data Consistency Replication Management Problems and Proposed Design Concepts Dependability Modeling for Replica Consistency GSPN Model for the Replica Consistency Estimation Markov Model Derived from GSPN Results Parameter Values Sensitivity Analyses for Replica Consistency Service Availability Analyses Conclusion and Outlook References Ten Fallacies of Availability and Reliability Analysis Prologue Basic Probability Theory Definitions Reliability Definitions Availability Definitions Fallacies ``Fault Tolerance Is an Availability Feature and Not a Reliability Feature'' ``Availability Is a Fraction While Reliability Is Statistical'' ``The Term `Software Reliability Growth' Is Unfortunate: Reliability Is Always a Non-increasing Function of Time'' ``MTTF Is the Whole Story about Reliability'' ``The Presence of Non-exponential Lifetime or Time-to-Repair Distributions Precludes Analytical Solution of State-Space Based R&A Models'' ``Availability Will Always Be Increased with More Redundancy'' ``Using Low-Cost Components Can Always Build Highly Available Systems'' ``A Ten-Times Decrease in MTTR Is Just as Valuable as a Ten-Times Increase in MTTF'' ``Improving Component MTTR Is the Key to Improve System Availability'' ``High-Availability Systems Should Have No Single Point-of-Failure'' Conclusions Analytical Availability Assessment of IT Services Introduction Related Work Reference Architecture and Fault Model Service Availability Assessment ICT-Service Mapping Calculating Availability of an E-Mail Service Total Service Availability Working with Incomplete Data ToolPrototype Conclusion References Author Index

This book constitutes the refereed proceedings of the 5th International Service Availability Symposium, ISAS 2008, held in Tokyo, Japan, in May 2008.

The 12 revised full papers presented together with 2 keynote papers and 2 tutorials were carefully reviewed and selected from 28 submissions. The papers are organized in topical sections on enterprise system dependability, software service availability, service availability platform, and service dependability analysis.