Large Scale Management of Virtual Machines Cooperative and Reactive Scheduling in Large-Scale Virtualized Platforms



Similar documents
Flauncher and DVMS Deploying and Scheduling Thousands of Virtual Machines on Hundreds of Nodes Distributed Geographically

An Experimental Study of Load Balancing of OpenNebula Open-Source Cloud Computing Platform

Autonomous and Energy-Aware Management of Large-Scale Cloud Infrastructures

The Eucalyptus Open-source Cloud Computing System

THE CC1 PROJECT SYSTEM FOR PRIVATE CLOUD COMPUTING

LSKA 2010 Survey Report I Device Drivers & Cloud Computing

Infrastructure as a Service (IaaS)

Keywords Distributed Computing, On Demand Resources, Cloud Computing, Virtualization, Server Consolidation, Load Balancing

FREE AND OPEN SOURCE SOFTWARE FOR CLOUD COMPUTING SERENA SPINOSO FULVIO VALENZA

Enabling Large-Scale Testing of IaaS Cloud Platforms on the Grid 5000 Testbed

Hadoop Distributed FileSystem on Cloud

Is Virtualization Killing SSI Research?

From Grid Computing to Cloud Computing & Security Issues in Cloud Computing

Infrastructure for Cloud Computing

Self-organization of applications and systems to optimize resources usage in virtualized data centers

How To Understand Cloud Computing

PRIVACY PRESERVATION ALGORITHM USING EFFECTIVE DATA LOOKUP ORGANIZATION FOR STORAGE CLOUDS

Is Virtualization Killing SSI Research?

VMM-Level Distributed Transparency Provisioning Using Cloud Infrastructure Technology. Mahsa Najafzadeh, Hadi Salimi, Mohsen Sharifi

A Distributed Approach to Dynamic VM Management

This is an author-deposited version published in : Eprints ID : 12902

Cloud Computing in Distributed System

Introduction to Cloud Computing

Review of Cloud Computing Architecture for Social Computing

Li Sheng. Nowadays, with the booming development of network-based computing, more and more

Dynamic Resource Allocation in Computing Clouds using Distributed Multiple Criteria Decision Analysis

A Comparison and Critique of Eucalyptus, OpenNebula and Nimbus

AN IMPLEMENTATION OF E- LEARNING SYSTEM IN PRIVATE CLOUD

Efficient Cloud Management for Parallel Data Processing In Private Cloud

SLA-aware Resource Scheduling for Cloud Storage

Keywords: Cloudsim, MIPS, Gridlet, Virtual machine, Data center, Simulation, SaaS, PaaS, IaaS, VM. Introduction

Cloud and Virtualization to Support Grid Infrastructures

Resource Scalability for Efficient Parallel Processing in Cloud

Virtual Machine Management with OpenNebula in the RESERVOIR project

IaaS Cloud Architectures: Virtualized Data Centers to Federated Cloud Infrastructures

Scheduler in Cloud Computing using Open Source Technologies

Comparison of Several Cloud Computing Platforms

I. INTRODUCTION. T Thiruvenkadam. V Karthikeyani

VIRTUAL RESOURCE MANAGEMENT FOR DATA INTENSIVE APPLICATIONS IN CLOUD INFRASTRUCTURES

An Open Source Solution for Virtual Infrastructure Management in Private and Hybrid Clouds

Elastic Management of Cluster based Services in the Cloud

GUEST OPERATING SYSTEM BASED PERFORMANCE COMPARISON OF VMWARE AND XEN HYPERVISOR

Exploring Resource Provisioning Cost Models in Cloud Computing

Unisys ClearPath Forward Fabric Based Platform to Power the Weather Enterprise

An Introduction to Virtualization and Cloud Technologies to Support Grid Computing

Two-Level Cooperation in Autonomic Cloud Resource Management

STeP-IN SUMMIT June 18 21, 2013 at Bangalore, INDIA. Performance Testing of an IAAS Cloud Software (A CloudStack Use Case)

Operating Systems Virtualization mechanisms

OpenNebula An Innovative Open Source Toolkit for Building Cloud Solutions

Migration of Virtual Machines for Better Performance in Cloud Computing Environment

Fault Tolerant Approaches in Cloud Computing Infrastructures

Unified API Governance in the New API Economy

Bare Metal Provisioning to OpenStack Using xcat

Cloud Computing from an Institutional Perspective

Multi Tenancy Access Control Using Cloud Service in MVC

Clouds Under the Covers. Elgazzar - CISC Fall

Cloud Computing and Open Source: Watching Hype meet Reality

CLEVER: a CLoud-Enabled Virtual EnviRonment

Security of Information Systems hosted in Clouds: SLA Definition and Enforcement in a Dynamic Environment

Science Clouds: Early Experiences in Cloud Computing for Scientific Applications Kate Keahey and Tim Freeman

An Implementation of Load Balancing Policy for Virtual Machines Associated With a Data Center

Deploying Business Virtual Appliances on Open Source Cloud Computing

Virtual Machine Monitors. Dr. Marc E. Fiuczynski Research Scholar Princeton University

Dynamic Resource allocation in Cloud

Virtualization Technologies (ENCS 691K Chapter 3)

Making a Smooth Transition to a Hybrid Cloud with Microsoft Cloud OS

Chapter 2 Addendum (More on Virtualization)

An Efficient Checkpointing Scheme Using Price History of Spot Instances in Cloud Computing Environment

Energetic Resource Allocation Framework Using Virtualization in Cloud

OpenNebula Leading Innovation in Cloud Computing Management

Comparative Study of Eucalyptus, Open Stack and Nimbus

The OpenNebula Standard-based Open -source Toolkit to Build Cloud Infrastructures

Eucalyptus LSS: Load-Based Scheduling on Virtual Servers Using Eucalyptus Private Cloud

Design and Building of IaaS Clouds

Cloud Optimize Your IT

CA ARCserve Replication and High Availability Deployment Options for Hyper-V

CA Cloud Overview Benefits of the Hyper-V Cloud

Transcription:

Large Scale Management of Virtual Machines Cooperative and Reactive Scheduling in Large-Scale Virtualized Platforms Adrien Lèbre EPI ASCOLA / HEMERA Flavien Quesnel, Phd Candidate February 2013

System Virtualization One to multiple OSes on a physical node thanks to a hypervisor (an operating system of OSes) Virtual Machines (VMs) Virtual Machine Monitor Hypervisor A virtual machine (VM) provides a faithful implementation of a physical processor s hardware running in a protected and isolated environment. Virtual machines are created by a software layer called the virtual machine monitor (VMM) that runs as a privileged task on a physical processor. Physical Machine (PM)

System Virtualization One to multiple OSes on a physical node thanks to a hypervisor (an operating system of OSes) Virtual Machines (VMs) Virtual Machine Monitor Hypervisor A virtual machine (VM) provides a faithful implementation of a physical processor s hardware running in a protected and isolated environment. Virtual machines are created by a software layer called the virtual machine monitor (VMM) that runs as a privileged task on a physical processor. Physical Machine (PM)

Context Efficient use of resources in data centers 0:00 12:00 24:00 0:00 12:00 24:00 Physical Machines 0:00 12:00 24:00 0:00 12:00 24:00 Virtual Machines Case 1: Static Consolidation Hypervisor Hypervisor Physical Machines 3

Context Efficient use of resources in data centers 0:00 12:00 24:00 0:00 12:00 24:00 Physical Machines 0:00 12:00 24:00 0:00 12:00 24:00 Virtual Machines Case 1: Static Consolidation Hypervisor Hypervisor Resources may not be fully used Physical Machines 3

Context Efficient use of resources data centers Schedule VMs according to their effective usage Physical Machines Virtual Machines Case 2: Dynamic Consolidation Hypervisor Physical Machine 4

Context Efficient use of resources data centers Schedule VMs according to their effective usage Physical Machines Virtual Machines Case 2: Dynamic Consolidation Hypervisor Physical Machine 4

Context Efficient use of resources data centers Schedule VMs according to their effective usage Physical Machines Virtual Machines Case 2: Dynamic Consolidation Hypervisor Physical Machine 4

Context Efficient use of resources data centers Schedule VMs according to their effective usage Physical Machines Virtual Machines Case 2: Dynamic Consolidation Hypervisor Hypervisor Physical Machine 4

Context Efficient use of resources data centers Schedule VMs according to their effective usage Physical Machines Virtual Machines Case 2: Dynamic Consolidation Hypervisor Hypervisor Physical Machine 4

Context Efficient use of resources data centers Schedule VMs according to their effective usage Physical Machines Virtual Machines Case 2: Dynamic Consolidation Hypervisor Physical Machine 4

First solutions... Centralized approaches Période 1. Resource 1. Collecte Monitoring des informations 3. Applying 3. Application reconfiguration de actions la décision 2. Computing 2. Prise a viable de scheduling décision 1 2 3 durée Time 5

First solutions... Centralized approaches Charge Load de travail Machine Virtual virtuelle Machine 1 Machine Virtual virtuelle Machine 2 Temps Time 1 2 3 Scalability/Reactivity concerns 5

Proposal overview Main characteristics Event driven Peer to peer, no service node Local interactions between nodes Monitoring Scheduling 6

DVMS Algorithm i 7

DVMS Algorithm i 7

DVMS Algorithm i L 7

DVMS Algorithm i L 7

DVMS Algorithm i L 7

DVMS Algorithm 7

DVMS Algorithm i 8

DVMS Algorithm i ml i 9

DVMS Algorithm i ml ml i 10

DVMS Algorithm: Shortcuts i ml i ml 11

DVMS Algorithm: Shortcuts i ml i ml 12

DVMS Algorithm: Shortcuts i ml i ml 13

Pros Reactivity/scalability Scheduling started when an event is generated Few nodes considered for scheduling much faster computation Resources are partitioned in time and space Several events can be processed simultaneously and independently First validation: SimGrid 100K VMs/10K PMs Second validation: in vivo experiments? 14

Large Scale Mamangent of VMs Investigating VM concerns implies to... Deploy the template Configure/Start each instance Control the execution... before conducting experiments Run Reserve Deploy OAR Reserve subnets G5ksubnets Reserve VLAN KaVLAN Kadeploy KaVLAN Reboot inside VLAN Physical nodes Virtual instances IP address assignment Boot Performing such a task on Few VMs on one node Hundred of VMs on one site Thousands of VMs on distinct sites Designing/Implementing tools to make the study and the investigation of virtualization concerns at large scale easier 15

Experiments with Flauncher DVMS prototype Language: Java Events activated: node overload Comparison between Entropy and DVMS Several non-viable configurations generated VMs started with Flauncher How Entropy and DVMS solve problems for each configuration? 16

Nodes reserved Nancy Sophia Nancy Rennes Sophia 27 116 141 224 52 2000 VMs, 251 nodes 3325 VMs, 309 nodes 17

Results Average time to solve all events (s) 70 Entropy DVMS 56 42 28 14 0 2000/251 3325/309 Experiment (nb of VMs / nb of nodes) 18

Results Average time to solve all events (s) 70 Entropy DVMS 12 Average time to solve an event (s) Entropy DVMS 56 42 28 14 9 6 3 0 2000/251 3325/309 Experiment (nb of VMs / nb of nodes) 0 2000/251 3325/309 Experiment (nb of VMs / nb of nodes) 18

Results Average time to solve an all events (s) (s) 1270 Entropy DVMS Average partition size to solve an event 400 Entropy DVMS 56 9 42 6 28 3 14 0 0 2000/251 3325/309 Experiment (nb of of VMs // nb of of nodes) 320 240 160 80 2.5 2.6 0 2000/251 3325/309 Experiment (nb of VMs / nb of nodes) 18

Lessons learnt Nodes not correctly deployed Redeploy, katapult Nodes that did not reboot correctly Kareboot Routing communications between sites KaVLAN Communications with more than 1024 machines Increase the size of ARP table Migration errors with libvirt Ensure that each node has a unique id... 19

Conclusion Flauncher, a set of scripts To create, stress and migrate a huge number of VMs distributed on several sites of Grid 5000 (10K VMs so far throughout 5 sites) That can be leveraged to study virtualization and live migration in large scale infrastructures (presented during the Grid 5000 winter school in Nantes, Dec 2012, [RR-8026]) Next Extension to execute advanced workflows (stress VMs, migrations,...) Collect./Analyze/Run real traces from cloud providers 20

Conclusion DVMS A distributed solution to schedule VMs dynamically in large scale infrastructures [VHPC 2011], [CCPE 2012] and an ongoing submission. Next Mid term, resiliency aspects ( Robustness of Large System of High Churn challenge) Long term, a building block of a proposal aiming at designing a P2P like Cloud OS (a co-supervised Phd between AVALON and ASCOLA) 21

Tutorials Flauncher https://www.grid5000.fr/mediawiki/index.php/ Booting_and_Using_Virtual_Machines_on_Grid'5000 DVMS http://www.emn.fr/z-info/ascola/doku.php? id=internet:members:fquesnel:deploysimulatorg5kv3 22

Thank you

References (1/2) [Cornabas 10] Jonathan Rouzaud Cornabas. A distributed and collaborative dynamic load balancer for virtual machine. In Euro-Par 2010 workshops, Ischia, Naples Italy, August 2010. Springer. [Feller et al. 11]" Eugen Feller, Louis Rilling, and Christine Morin. Snooze: A Scalable and Autonomic Virtual Machine Management Framework for Private Clouds. Research report, INRIA Rennes, Rennes, France, December 2011. [Hermenier et al. 09] Fabien Hermenier, Xavier Lorca, Jean M. Menaud, Gilles Muller, and Julia Lawall. Entropy: a consolidation manager for clusters. In VEE 09: Proceedings of the 2009 ACM SIGPLAN/SIGOPS international conference on Virtual execution environments, pages 41 50, New York, NY, USA, March 2009. ACM. [Hoffa et al. 08] On the Use of Cloud Computing for Scientific Workflows, Hoffa, C., G. Mehta, T. Freeman, E. Deelman, K. Keahey, B. Berriman, J. Good. SWBES 2008, Indianapolis, IN. December 2008 [Lèbre et al. 11] DISCOVERY, Beyond the Clouds, Lebre, A. and Anedda, P. and Gaggero, M. and Quesnel, F. In Euro-Par 2011 workshops, Bordeaux, France, August 2011. Springer. 24

References (2/2) [Lowe 09] Scott Lowe. Introducing VMware vsphere 4. Wiley Publishing Inc., Indianapolis, Indiana, first edition, September 2009. [Mastroiani et al. 11] Carlo Mastroianni, Michela Meo and Giuseppe Papuzzo. Self-economy in cloud data centers: statistical assignment and migration of virtual machines. In Euro-Par'11: Proceedings of the 17th international conference on Parallel processing, pages 407-418, Bordeaux, France, August 2011. Springer. [Nurmi et al. 09] Daniel Nurmi, Rich Wolski, Chris Grzegorczyk, Graziano Obertelli, Sunil Soman, Lamia Youseff, and Dmitrii Zagorodnov. The eucalyptus open-source cloud-computing system. In CCGRID 09: Proceedings of the 2009 9th IEEE/ACM International Symposium on Cluster Computing and the Grid, volume 0, pages 124 131, Washington, DC, USA, May 2009. IEEE Computer Society. [Sotomayor et al. 09] Borja Sotomayor, Rubén S. Montero, Ignacio M. Llorente, and Ian Foster. Virtual infrastructure management in private and hybrid clouds. IEEE Internet Computing, 13(5):14 22, September 2009. [Yagiz et al. 10] Yagiz O. Yazir, Chris Matthews, Roozbeh Farahbod, Stephen Neville, Adel Guitouni, Sudhakar Ganti, and Yvonne Coady. Dynamic resource allocation in computing clouds using distributed multiple criteria decision analysis. In Cloud 10: IEEE 3rd International Conference on Cloud Computing, pages 91 98, Los Alamitos, CA, USA, July 2010. IEEE Computer Society. 25