Computer Performance Engineering

Name: Computer Performance Engineering | 8th European Performance Engineering Workshop, EPEW 2011, Borrowdale, The English Lake District, October 12-13,2011, Proceedings
Brand: Springer
Price: 53.49 EUR
Availability: OnlineOnly

8th European Performance Engineering Workshop, EPEW 2011, Borrowdale, The English Lake District, October 12-13,2011, Proceedings

Nigel Thomas(Herausgeber*in)

Springer (Verlag)

Erschienen am 28. Januar 2012

XII, 341 Seiten

E-Book

PDF mit Wasserzeichen-DRM

Systemvoraussetzungen

978-3-642-24749-1 (ISBN)

53,49 €inkl. 7% MwSt.

Systemvoraussetzungen

für PDF mit Wasserzeichen-DRM

E-Book Einzellizenz

Als Download verfügbar

Beschreibung

Weitere Details

Weitere Ausgaben

Inhalt

Intro
Title
Preface
Organization
Table of Contents
Invited Papers
Performance Engineering for Cloud Computing
Evolution of Performance Engineering
Cloud Context
Motivation of Performance Challenges
Classical Performance Engineering
Cloud Specific Challenges
SAAS Performance Design
Autonomics
Performance Monitoring
Resource Utilization
Data Analysis
Conclusions
References
Engineering of Self-aware IT Systems and Services: State-of-the-Art and Research Challenges
References
Regular Papers
Accommodating Short and Long Web Traffic Flows over a DiffServ Architecture
Introduction
Long Flow Promotion (LFP)
Preferential Treatment for Short Flows (S1)
Packets Promotion Close to the Threshold (S2)
Detecting Elephant Flows (S3)
Scheduling the Packets
Simulation Results
Conclusions
References
Automatic Synchronisation Detection in Petri Net Performance Models Derived from Location Tracking Data
Introduction
Synchronisation Detection Mechanism
Algorithm Description
Synchronisation Representation in Our Models
Case Study
Results
Conclusion
References
Performance Evaluation of Business Processes through a Formal Transformation to SAN
Introduction
Related Works
Fundamentals
Stochastic Automata Network
Business Process Model and Notation
Definitions for the BPMN to SAN Conversion
BPMN Graph Definitions
SAN Model Definitions
The Conversion Algorithm
Example - Shipment Process of a Hardware Retailer
Concluding Remarks
References
Monotonicity and Efficient Computation of Bounds with Time Parallel Simulation
Introduction
Time Parallel Simulation with Fix-Up Phases
Monotone Systems, Comparison of States and Comparison of Sequences
Fast Parallel Computation of Bounds with TPS
Modelling a Web Server
Conclusion
References
Stochastic Restricted Broadcast Process Theory
Introduction
Evaluation Factors
Stochastic RBPT
Actions: Types and Rates
Syntax
Execution Mechanisms and Formal Semantics
From Configuration to CTMC
Weak Markovian Network Bisimilarity
Markovian Semantics of MANETs
A Leader Election Algorithm for MANETs
Protocol Specification
Protocol Analysis
Conclusion and Future Work
References
Higher Moment Analysis of a Spatial Stochastic Process Algebra
Introduction
Markovian Agent Models
Higher Moment ODEs for the Master Equation
Higher (joint) Moments in MAMs
Markovian Agent Spatial Stochastic Process Algebra
A Simple MASSPA Example
Translating MASSPA into Mass Action Type Reactions
Worked Example: A Simplified Spatial WSN
Conclusion
References
Optimization for Multi-thread Data-Flow Software
Introduction
Related Work
Analytical Model
Optimization Approaches
Optimization Algorithm
Real Measurements
Experiments
Verification of the Analytical Approach
Removing the Bottleneck
Conclusion
References
Formal Mapping of WSLA Contracts on Stochastic Models
Introduction
Background
Stochastic Discrete Event System (SDES)
Web Service Level Agreement (WSLA)
Outline of the Mapping Process
Step 1: Formal Definition of Basic WSLA Semantics
Step 2: MeasurementDirective(s) Mapping
StatusRequest and Status
InvocationCount
Gauge
Counter
ResponseTime
DownTime
Step 3: Schedule Mapping
Step 4: Function(s) Mapping
Step 5: SLO Result
Case Study
System Description
Mapping WSLA to SDES
Discussion
Conclusion
References
Comparison of the Mean-Field Approach and Simulation in a Peer-to-Peer Botnet Case Study
Introduction
Model of the Botnet Behavior
Mean-Field Approximation
Results
Exploiting the Speed-Up
Variations of the Method
Conclusion
References
WMTools - Assessing Parallel Application Memory Utilisation at Scale
Introduction
Related Work
The WMTrace Library and Analysis Framework
Stage 1: Memory Event Tracing
Stage 2: Post-Execution Trace Analysis
Case Study
High Water Mark
Memory Usage over Time
Tool Overhead and Comparison
Compression
Analysis
Conclusions
References
On Stochastic Fault-Injection for IP-Packet Loss Emulation
Introduction
Measurement, Modelling, and Injection of IP Packet Loss
Packet-Loss Models
Stochastic Fault-Injection
The Problem with Markovian Models in Fault-Injection
Scenarios for Fault Injection
Modules for Fault Injection
Experiments
Scenarios
Results
Concluding Discussion
References
Analysis of Gossip-Based Information Propagation in Wireless Mesh Networks
Introduction
Shuffle Protocol
Description
Probabilistic State Transition System
WMN Model of the Shuffle Protocol
Lower Bound on rp(t+1)
Modeling Replication and Coverage Ratio with Differential Equations
Evaluation of the Model
Conclusion and Future Work
References
Multi-class Network with Phase Type Service Time and Group Deletion Signal
Introduction
Preliminaries
Definition of Quasi-reversibility of Chao, Miyazawa and Pinedo
Network of Quasi-reversible Queues with Signals and Instantaneous Movement
A LIFO Multi-class PH Queue with Signal Deletion Customers of Same Sub-class
Description
Stationary Distribution
Quasi-reversibility
Network of LIFO Multi-class PH Queues with Negative Signal and Group-Deletion Signal
Description
Stationary Distribution
Concluding Remarks
References
Critical Level Policies in Lost Sales Inventory Systems with Different Demand Classes
Introduction
Problem Formulation
Model Description
Optimal Control Problem
Critical Level Policies
Hypoexponential Service
Model Description
Increasing Property
Convexity
Critical Level Policy
Submodularity
Monotone Switching Curve
Model Extensions
Hyperexponential Service
Adding Holding Costs
Conclusions
References
Model-Based Evaluation and Improvement ofPTP Syntonisation Accuracy in Packet-SwitchedBackhaul Networks for Mobile Applicatio
Introduction
Technical Background
Accurate and Efficient Simulation of Backhaul Networks
PDV Characteristics
Simulation Models for PDV Evaluation
Validation
Delay-Step Elimination
Analytical Evaluation of PDV in Tree-Structured Networks
Closed-Form Expression for p2pPDV
Reducing PDV in Tree-Structured Networks
Evaluation
Discussion
Conclusion
References
Light-Weight Parallel I/O Analysis at Scale
Introduction
Related Work
Warwick RIOT
Case Study
Input/Output Benchmarks
MPI-IO and POSIX Bandwidth Tracing
Collective Buffering in ROMIO
Analysis of HDF-5
Analysis of PLFS Middleware
Conclusions
Future Work
References
Poster Presentations
Can Linear Approximation Improve Performance Prediction ?
Introduction
Performance Models and Synthetic Applications
Random Software Applications
Queueing Petri Net Models
Determining Maximum Accuracy
Perfect Service Demand Adjustment
Linear Service Demand Adjustment
Assessing Realistic Accuracy
Limitations
Conclusion
References
TWOEAGLES: A Model Transformation Tool from Architectural Descriptions to Queueing Networks
Introduction
The Transformation from Æmilia to Queueing Networks
The Transformation Source: Æmilia
The Transformation Target: Queueing Networks
The Transformation at a Glance
A Hierarchical Approach to the Transformation
The Eclipse Plugin Æmilia_to_QN
The Architecture of TwoEagles
TwoEagles at Work: An Automated Teller Machine
Conclusions and Future Work
References
A Tool Suite for Modelling Spatial Interdependencies of Distributed Systems with Markovian Agents
Introduction
Related Works
Markovian Agents
Algorithmic Generation of Spatial Dependencies
The Image-Based Tools
The Graph-Based Tool
Computing the Number of Hops in Critical Infrastructures with MA
Performance Evaluation
Numerical Results
Conclusions
References
A Grid Broker Pricing Mechanism for Temporal and Budget Guarantees
Introduction
The Pricing Mechanism
Experimental Setup
Evaluation
Related Work
Conclusion and Future Work
References
Visualisation for Stochastic Process Algebras: The Graphic Truth
Introduction
Modelling in PEPA
The Argument for Visualisation
Visualisation of PEPA Models
Constructing Performance Properties
Implementation Details
Related Work
Conclusions
References
Efficient Experiment Selection in Automated Software Performance Evaluations
Introduction
Motivation and Overview
Software Performance Curves
Integrating Software Performance Curves and Model-Driven Performance Analyses
Related Work
Statistical Model Inference
Kriging
MARS
Experiment Selection
Random Experiment Selection Based on Global Prediction Error
Equidistant Experiment Selection Based on Global Prediction Error
Adaptive Experiment Selection Based on Local Prediction Error
Case Study and Validation
Communication Server Case Study
Enterprise Application Case Study
Discussion
Summary
References
Author Index

Systemvoraussetzungen

Als PDF speichern Als Link merken