1 Agility in Virtualized Utility Computing Hangwei Qian, Elliot Miller, Wei Zhang Michael Rabinovich, Craig E. Wills {EECS Department, Case Western Reserve.

Slides:



Advertisements
Similar presentations
Elastic Provisioning In Virtual Private Clouds
Advertisements

1 Efficient Application Placement in a Dynamic Hosting Platform Zakaria Al-Qudah – CWRU Hussein Alzoubi – CWRU Mark Allman – ICSI Michael Rabinovich –
Capacity Planning in a Virtual Environment
High Availability Deep Dive What’s New in vSphere 5 David Lane, Virtualization Engineer High Point Solutions.
© 2008 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice OpenVMS Guests Under HPVM Barry Kierstein.
SLA-Oriented Resource Provisioning for Cloud Computing
System Center 2012 R2 Overview
The Case for Enterprise Ready Virtual Private Clouds Timothy Wood, Alexandre Gerber *, K.K. Ramakrishnan *, Jacobus van der Merwe *, and Prashant Shenoy.
The Who, What, Why and How of High Performance Computing Applications in the Cloud Soheila Abrishami 1.
Xen , Linux Vserver , Planet Lab
Copyright © 2005 VMware, Inc. All rights reserved. Optimizing Blade Deployment with Virtual Infrastructure Eric Horschman Group Product Marketing Manager.
Virtualization and Cloud Computing Virtualization David Bednárek, Jakub Yaghob, Filip Zavoral.
11 HDS TECHNOLOGY DEMONSTRATION Steve Sonnenberg May 12, 2014 © Hitachi Data Systems Corporation All Rights Reserved.
Post-Copy Live Migration of Virtual Machines Michael R. Hines, Umesh Deshpande, Kartik Gopalan Computer Science, Binghamton University(SUNY) SIGOPS 09’
Introduction to Virtualization
Transform your desktop with virtualization. 22 Agenda Evolution of VDI VDI Solution VDI Use Cases Questions & Answers.
Virtualization on the Intel Platform Scott Elliott Senior Systems Network Specialist Christie Digital A Customer Implementation with VMware and IBM.
Keith Wiles DPACC vNF Overview and Proposed methods Keith Wiles – v0.5.
OnCall: Defeating Spikes with Dynamic Application Clusters Keith Coleman and James Norris Stanford University June 3, 2003.
U NIVERSITY OF M ASSACHUSETTS, A MHERST Department of Computer Science Virtualization in Data Centers Prashant Shenoy
INTRODUCTION TO CLOUD COMPUTING CS 595 LECTURE 4.
Virtualization for Cloud Computing
Copyright Tim Antonowicz, This work is the intellectual property of the author. Permission is granted for this material to be shared for non- commercial,
VMware vCenter Server Module 4.
ProjectWise Virtualization Kevin Boland. What is Virtualization? Virtualization is a technique for deploying technologies. Virtualization creates a level.
Virtualization Performance H. Reza Taheri Senior Staff Eng. VMware.
To run the program: To run the program: You need the OS: You need the OS:
VMware vSphere 4 Introduction. Agenda VMware vSphere Virtualization Technology vMotion Storage vMotion Snapshot High Availability DRS Resource Pools Monitoring.
Copyright © 2010 Platform Computing Corporation. All Rights Reserved.1 The CERN Cloud Computing Project William Lu, Ph.D. Platform Computing.
Networking Virtualization Using FPGAs Russell Tessier, Deepak Unnikrishnan, Dong Yin, and Lixin Gao Reconfigurable Computing Group Department of Electrical.
Dual Stack Virtualization: Consolidating HPC and commodity workloads in the cloud Brian Kocoloski, Jiannan Ouyang, Jack Lange University of Pittsburgh.
How to Resolve Bottlenecks and Optimize your Virtual Environment Chris Chesley, Sr. Systems Engineer
SAIGONTECH COPPERATIVE EDUCATION NETWORKING Spring 2010 Seminar #1 VIRTUALIZATION EVERYWHERE.
SAIGONTECH COPPERATIVE EDUCATION NETWORKING Spring 2009 Seminar #1 VIRTUALIZATION EVERYWHERE.
Virtual Machine Course Rofideh Hadighi University of Science and Technology of Mazandaran, 31 Dec 2009.
Efficient Application Placement in a Dynamic Hosting Platform Zakaria Al-Qudah, Hussein A. Alzoubi, Mark Allman, Michael Rabinovich, Vincenzo Liberatore.
Appendix B Planning a Virtualization Strategy for Exchange Server 2010.
Copyright © 2011 EMC Corporation. All Rights Reserved. MODULE – 6 VIRTUALIZED DATA CENTER – DESKTOP AND APPLICATION 1.
VMware Infrastructure 3 The Next Generation in Virtualization.
Improving Network I/O Virtualization for Cloud Computing.
Virtualization in the Data Center Virtual Servers – How it works – Pros – Cons IPAC’s implementation – Hardware resource usage and trends – Virtualization.
COMS E Cloud Computing and Data Center Networking Sambit Sahu
Our work on virtualization Chen Haogang, Wang Xiaolin {hchen, Institute of Network and Information Systems School of Electrical Engineering.
Eric Burgener VP, Product Management A New Approach to Storage in Virtual Environments March 2012.
Dynamic and Secure Application Consolidation with Nested Virtualization and Library OS in Cloud Kouta Sannomiya and Kenichi Kourai (Kyushu Institute of.
Virtual Machines Created within the Virtualization layer, such as a hypervisor Shares the physical computer's CPU, hard disk, memory, and network interfaces.
1 Grid Activity Summary » Grid Testbed » CFD Application » Virtualization » Information Grid » Grid CA.
Copyright © 2005 VMware, Inc. All rights reserved. How virtualization can enable your business Richard Allen, IBM Alliance, VMware
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Capacity Planning in a Virtual Environment Chris Chesley, Sr. Systems Engineer
REMINDER Check in on the COLLABORATE mobile app Best Practices for Oracle on VMware - Deep Dive Darryl Smith Chief Database Architect Distinguished Engineer.
© ExplorNet’s Centers for Quality Teaching and Learning 1 Explain the purpose of Microsoft virtualization. Objective Course Weight 2%
A Measured Approach to Virtualization Don Mendonsa Lawrence Livermore National Laboratory NLIT 2008 by LLNL-PRES
© 2012 Eucalyptus Systems, Inc. Cloud Computing Introduction Eucalyptus Education Services 2.
Architecture of a platform for innovation and research Erik Deumens – University of Florida SC15 – Austin – Nov 17, 2015.
VIRTUAL MACHINE – VMWARE. VIRTUAL MACHINE (VM) What is a VM? – A virtual machine (VM) is a software implementation of a computing environment in which.
UFIT Infrastructure Self-Service. Service Offerings And Changes Virtual Machine Hosting Self service portal Virtual Machine Backups Virtual Machine Snapshots.
1 Automated Power Management Through Virtualization Anne Holler, VMware Anil Kapur, VMware.
The Docker Container Approach to Build Scalable and Performance Testing Environment Pankaj Rodge, VMware.
Unix Server Consolidation
Bentley Systems, Incorporated
Virtualization OVERVIEW
Kenichi Kourai Kouta Sannomiya Kyushu Institute of Technology, Japan
R
Virtual Servers.
Oracle Solaris Zones Study Purpose Only
1. 2 VIRTUAL MACHINES By: Satya Prasanna Mallick Reg.No
Managing Clouds with VMM
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Presentation transcript:

1 Agility in Virtualized Utility Computing Hangwei Qian, Elliot Miller, Wei Zhang Michael Rabinovich, Craig E. Wills {EECS Department, Case Western Reserve University} {CS Department, Worcester Polytechnic Institute}

2 Internet Applications: Resource Provisioning Challenge  Unpredictable demand  Challenge: how much resource to provision?  Too little – lose business  Too much – poor utilization  Utility Computing is a promising answer

3 Resource Management Alternatives  Application server sharing  Different applications running on the same app server  OS-level sharing  Each application runs in dedicated app server  Several app servers share physical machine  Dedicated physical machines  Each physical machine runs one app server with one app  Resources are assigned at the granularity of entire machines  Isolation (security, fault, performance) vs. utilization tradeoff

4 An Appealing Option: Virtual Machines  Almost as good isolation as physical machines reassignment  Much better utilization than physical machine reassignment  Heterogeneity support

5 Target Utility Computing Platform Applications run on dedicated app servers App servers run on dedicated VMs App instances form app-level clusters

6 Goal  Improve the agility of Utility Computing “how quickly Utility Computing Platform can react to changing demands.”

7 VM-based resource management alternatives  Migration of VMs  Redeployment of Application Servers on Demand

8 Migration of VMs  Start/Stop VMs  Predeploy everywhere  Start more instances in high-demand areas  Stop underloaded instances  Clone VM  Stop overloaded VMs on overutilized hosts  Copy stopped VMs to other underutilized hosts  Start the cloned VM on target hosts  Suspend/resume  Pre-deploy VMs across Hosts  Suspend VMs with Unpopular Applications  Resume VMs with popular Applications  Live VM Migration

9 Redeployment of Application Servers on Demand  Pre-deploy VMs across Hosts  Remove app servers with Unpopular Applications from VMs  Add app servers with popular Applications to VMs

10 Our Approach: Ghost VMs  Active VM: alive and utilized for application request processing  Ghost VM: alive but unutilized Fully participate in cluster maintenance Do not process any requests Spend little CPU cycles

11 Resource Reassignment Mechanism Ghost implementation options: switch reconfiguration on demand -Remove VMs from switch -Set the Max Connection option to minimum (1) for the ghost VMs

12 Evaluation  Experiment Environment  Hardware: Intel (R) 4-core 2.33 GHz, CPU, 4G memory, 146G disk with 15K RPM Nortel Alteon 2208 Application Switch  Software: Linux generic SMP VMware Server Websphere 6 Network Deployment

13 Migration of VMs  Start/Stop VM  About 64s to start a VM when its memory is flushed out, with a competing VM allocating memory moderately  About 20s to start the cluster node agent  About 97s to start the application server  Totally about 181s  Clone VMs  About 15 minutes for a VM with 1G memory and 10G disk in a 100Mbps local network.

14 Migration of VMs (cont’ed)  Suspend/Resume VMs  About 5s on average in a repeated experiments  About 10s on average for VM with memory swapped out  About 14s on average with a competing VM (more realistic)  About 30s for app server on the resumed VM to get re- integrated into the cluster  Totally about 44s

15 Redeployment of Application Servers on Demand  About 95s to stop a cluster member (application server)  About 19s to create a cluster member  About 97s to start a cluster member  Totally about 211s

16 Agility of the Above Methods MethodAgility (sec) Migration of VMs~180 Redeployment of App Servers~210 Suspend/Resume VMs>44  The best agility we can expect is: 44s

17 Active/Ghost VMs  Overhead of ghost VM: 1.4%-3.3% CPU  Agility  7s: Add/remove VM from switch  2s: Set the Max connection to minimum(1)

18 Memory Swap-in Problem?  For a VM with 1G memory  about 652s to swap in all the memory  about 267s to swap in 50% memory  about 80s to swap in 5% memory  Solution: Keep Ghosts in memory!  Only run as many active+ghost VMs as total physical memory allows  Also reserve memory for host OS (1G) Adding memory is much cheaper than adding physical machines Only very small number of ghost VMs are needed to hide resource reassignment latency  VMware ESX: Reserved Memory for VMs  VM state hierarchy  Suspended  Ghost  Active

19 Evaluation Summary MethodAgility (sec) Migration of VMs~180 Redeployment of App Servers~210 Suspend/Resume VMs>44 Active/Ghost VMs2-7  Active/Ghost VMs has the best Agility  The price is memory overhead  Summary

20 Future Work  Extend the scale of our test platform  Move to VMware ESX  More investigation of resource tradeoffs for ghost, suspended and active VMs  Explore more on global and local data center resource management  Build full-function prototype for resource management in Utility Computing

21 Summary  Internet applications introduce resource provisioning challenge  Utility computing is a promising paradigm  VMs gain popularity as resource management approach  Agility is an important characteristic of a utility computing platform  Ghost VMs as a mechanism for agile utility computing

22 Thank you ! Questions?