Budapesti Műszaki és Gazdaságtudományi Egyetem Méréstechnika és Információs Rendszerek Tanszék Infrastructure for Model-based Control of Distributed IT.

Slides:



Advertisements
Similar presentations
This course is designed for system managers/administrators to better understand the SAAZ Desktop and Server Management components Students will learn.
Advertisements

© 2006 OpenGridForum Craig Lee, President, OGF The Larger Context for Green IT Systems OGF-25 A non-profit, federally funded R&D center.
AMUSE Autonomic Management of Ubiquitous Systems for e-Health Prof. J. Sventek University of Glasgow In collaboration.
Database Architectures and the Web
CS 795 – Spring  “Software Systems are increasingly Situated in dynamic, mission critical settings ◦ Operational profile is dynamic, and depends.
© Neeraj Suri EU-NSF ICT March 2006 Budapesti Műszaki és Gazdaságtudományi Egyetem Méréstechnika és Információs Rendszerek Tanszék Zoltán Micskei
C LOUD C OMPUTING Presented by Ye Chen. What is cloud computing? Cloud computing is a model for enabling ubiquitous, convenient, on- demand network access.
DataGrid is a project funded by the European Union 22 September 2003 – n° 1 EDG WP4 Fabric Management: Fabric Monitoring and Fault Tolerance
Variability Oriented Programming – A programming abstraction for adaptive service orientation Prof. Umesh Bellur Dept. of Computer Science & Engg, IIT.
1 Sensor Networks and Networked Societies of Artifacts Jose Rolim University of Geneva.
Documenting the Existing Network - Starting Points IACT 418 IACT 918 Corporate Network Planning.
Rutgers PANIC Laboratory The State University of New Jersey Self-Managing Federated Services Francisco Matias Cuenca-Acuna and Thu D. Nguyen Department.
The Bio-Networking Architecture: An Infrastructure of Autonomic Agents in Pervasive Networks Jun Suzuki netresearch.ics.uci.edu/bionet/
© 2008 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice Automated Workload Management in.
1 Exploring Data Reliability Tradeoffs in Replicated Storage Systems NetSysLab The University of British Columbia Abdullah Gharaibeh Matei Ripeanu.
CERN - IT Department CH-1211 Genève 23 Switzerland t Oracle and Streams Diagnostics and Monitoring Eva Dafonte Pérez Florbela Tique Aires.
Maintaining and Updating Windows Server 2008
MCTS Guide to Microsoft Windows Server 2008 Network Infrastructure Configuration Chapter 11 Managing and Monitoring a Windows Server 2008 Network.
Adaptive Server Farms for the Data Center Contact: Ron Sheen Fujitsu Siemens Computers, Inc Sever Blade Summit, Getting the.
Towards Autonomic Hosting of Multi-tier Internet Services Swaminathan Sivasubramanian, Guillaume Pierre and Maarten van Steen Vrije Universiteit, Amsterdam,
Cloud Attributes Business Challenges Influence Your IT Solutions Business to IT Conversation Microsoft is Changing too Supporting System Center In House.
Budapesti Műszaki és Gazdaságtudományi Egyetem Méréstechnika és Információs Rendszerek Tanszék Scheduling in Windows Zoltan Micskei
Generating Adaptation Policies for Multi-Tier Applications in Consolidated Server Environments College of Computing Georgia Institute of Technology Gueyoung.
1 Exploring Data Reliability Tradeoffs in Replicated Storage Systems NetSysLab The University of British Columbia Abdullah Gharaibeh Advisor: Professor.
Database Architectures and the Web
Report : Zhen Ming Wu 2008 IEEE 9th Grid Computing Conference.
1 NETE4631 Managing the Cloud and Capacity Planning Lecture Notes #8.
The Autonomic Cloud An ASCENS case study Future Emerging Technologies.
Virtualization. Virtualization  In computing, virtualization is a broad term that refers to the abstraction of computer resources  It is "a technique.
Adaptability for flexible mobile service provision in 3G and beyond Nikos Houssos
Active Monitoring in GRID environments using Mobile Agent technology Orazio Tomarchio Andrea Calvagna Dipartimento di Ingegneria Informatica e delle Telecomunicazioni.
Computer Science Open Research Questions Adversary models –Define/Formalize adversary models Need to incorporate characteristics of new technologies and.
Cluster Reliability Project ISIS Vanderbilt University.
Prof. N. P. Pathak - Dept. of I.T.1 Unit 4 Inventory Management Process OSS Essentials by Kornel Terplan.
Management for IP-based Applications Mike Fisher BTexaCT Research
P-1 © 2005 NeuralWare. All rights reserved. Using Neural Networks in Decision Support Systems Introduction Core Technology Building and Deploying Neural.
Managing the Oracle Application Server with Oracle Enterprise Manager 10g.
SOFTWARE DESIGN AND ARCHITECTURE LECTURE 05. Review Software design methods Design Paradigms Typical Design Trade-offs.
Handling Session Classes for Predicting ASP.NET Performance Metrics Ágnes Bogárdi-Mészöly, Tihamér Levendovszky, Hassan Charaf Budapest University of Technology.
Ó 1998 Menascé & Almeida. All Rights Reserved.1 Part V Workload Characterization for the Web.
Look, Ma, No Hardware -Stephanie Schossow. Cisco & VMware  September 16, Industry leaders in virtualization Cisco and VMware® announced that they.
Creating SmartArt 1.Create a slide and select Insert > SmartArt. 2.Choose a SmartArt design and type your text. (Choose any format to start. You can change.
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Module 9 Planning and Implementing Monitoring and Maintenance.
June 13-15, 2007Policy 2007 Infrastructure-aware Autonomic Manager for Change Management H. Abdel SalamK. Maly R. MukkamalaM. Zubair Department of Computer.
Managing Web Server Performance with AutoTune Agents by Y. Diao, J. L. Hellerstein, S. Parekh, J. P. Bigus Presented by Changha Lee.
Data Communications and Networks Chapter 9 – Distributed Systems ICT-BVF8.1- Data Communications and Network Trainer: Dr. Abbes Sebihi.
Aneka Cloud ApplicationPlatform. Introduction Aneka consists of a scalable cloud middleware that can be deployed on top of heterogeneous computing resources.
Performance Testing Test Complete. Performance testing and its sub categories Performance testing is performed, to determine how fast some aspect of a.
Module 14 Monitoring and Maintaining Windows Server® 2008 Servers.
CERN IT Department CH-1211 Genève 23 Switzerland t CERN IT Monitoring and Data Analytics Pedro Andrade (IT-GT) Openlab Workshop on Data Analytics.
Control-Theoretic Approaches for Dynamic Information Assurance George Vachtsevanos Georgia Tech Working Meeting U. C. Berkeley February 5, 2003.
IHP Im Technologiepark Frankfurt (Oder) Germany IHP Im Technologiepark Frankfurt (Oder) Germany ©
Cyberinfrastructure Overview of Demos Townsville, AU 28 – 31 March 2006 CREON/GLEON.
By Harshal Ghule Guided by Mrs. Anita Mahajan G.H.Raisoni Institute Of Engineering And Technology.
Maintaining and Updating Windows Server 2008 Lesson 8.
SESM Demonstrator FPGA Power Node Prototype Emilio Bisbiglio, SESM, Przemyslaw Osocha, SESM,
RESERVOIR Service Manager NickTsouroulas Head of Open-Source Reference Implementations Unit Juan Cáceres
Virtual Data Center LAN
Understanding the New PTC System Monitor (PSM/Dynatrace) Application’s Capabilities and Advanced Usage Stephen Vaillancourt PTC Technical Support –Technical.
Regional Operations Centres Core infrastructure Centres
Introduction.
Operating Systems : Overview
VMware vRealize® Operations™ Management Pack for Pure Storage
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Cloud Management Mechanisms
Operating Systems : Overview
write on board in advance: handouts, names, Hoare quote
Presentation transcript:

Budapesti Műszaki és Gazdaságtudományi Egyetem Méréstechnika és Információs Rendszerek Tanszék Infrastructure for Model-based Control of Distributed IT systems Gergely János Paljak Advisors: András Pataricza, Tamás Kovácsházy

Federated, virtualized data center Virtualization Intelligent system management Application A Application B Server 1 Server 2 Server 3 Central resource pools Individual servers, static allocation sensors for capacity/utilization monitoring Individual servers, static allocation sensors for capacity/utilization monitoring Virtual servers: Easier reconfiguration in runtime Virtual servers: Easier reconfiguration in runtime Federated DC: centralized management, intelligent decision making for provisioning

Manual configuration of system supervision Software Component Monitoring Server Presentation Server Historical Data Deployment of supervisory infrastructure: -Moderate effort -Low probability of faults Deployment of supervisory infrastructure: -Moderate effort -Low probability of faults S S S S S S Deployment of sensors (local agents/agentless): -Moderate effort -Some configuration faults can be introduced Deployment of sensors (local agents/agentless): -Moderate effort -Some configuration faults can be introduced Configuring alerting & diagnostic logic: -Effort larger by orders of magnitude -Faulty configurations due to ad-hoc design Configuring alerting & diagnostic logic: -Effort larger by orders of magnitude -Faulty configurations due to ad-hoc design

System Management as a Control Problem Software Component Service deployed on provides Decision Making Control theory Monitoring Provisioning Controlled Plant Sensors Controller Actuator Collect and store data about the state of the infrastucture Based on human expertise or automation Effectuate changes in the infrastructure applied to IT Infrastructures Control Objective (e.g. SLA) Control Policy Supervised NodeMonitoring / Control Node

Architecture Realistic infrastructure: Multi-tier Widely-used components Realistic infrastructure: Multi-tier Widely-used components Realistic workload Reconfigurable in runtime Integrated system monitoring, wide range of measured metrics Integrated intelligent data processing (in MatLab)

Measured attributes  We chose to measure all possibly relevant performance attributes  Leave the selection of truly relevant attributes to data processing PlatformPlatform AgentAgent Processes AgentAgentAgentAgent AgentAgentAgentAgent Middle- ware Clients Ex. CPU idle (%), free memory (kb), network bytes sent Ex. MySQL threads, Tomcat processing time, Apache open connections Ex. MySQL threads, Tomcat processing time, Apache open connections

Variable selection - the dilemma LinearEntropy based Objectivemin E(distance error 2 )max (shape similarity) Feature preservationSimple projectionMore context InvarianceLinear transformationAny bijective function Main characteristics retainedAvarege distanceShape Plane mirror Less details Less distortion Plane mirror Less details Less distortion Spheric mirror More details Huge distortion Spheric mirror More details Huge distortion Paljak, Kocsis, Égel, Tóth, Pataricza: „Sensor Selection for IT infrastructure Monitoring”, AUTONOMICS ‘09

Future work metric Software Component Service deployed on provides metric Prediction Control Actuator (reconfiguration) Actuator (reconfiguration) Dimension reduction Prediction based on the smaller state space Model-predictive control for proactivity Adaptive reconfiguration

Summary  An infrastructure for evaluating control methods in system management was built  The pilot infrastructure is a three-tier server system with integrated system monitoring and data processing  We are going to use this infrastructure for o Modeling multi-tier systems (identification) o Create algorithms for control o Evaluate the developped algorithms