Resource Allocation in Virtual Desktop Clouds: VMLab-GENI Experiment Rohit Patali, Prasad Calyam, Mukundan Sridharan, Alex Berryman The Ohio State University,

Slides:



Advertisements
Similar presentations
OnTimeMeasure Integration with Gush Prasad Calyam, Ph.D. (PI) Paul Schopis, (Co-PI) Tony Zhu (Software Programmer) Alex Berryman (REU Student)
Advertisements

Performance Testing - Kanwalpreet Singh.
FIBRE-BR Meeting GENI I&M Marcelo Pinheiro. Agenda GENI Overview GENI User groups GENI I&M Use Cases GENI I&M Services.
SLA-Oriented Resource Provisioning for Cloud Computing
System Center 2012 R2 Overview
GENI Experiment Control Using Gush Jeannie Albrecht and Amin Vahdat Williams College and UC San Diego.
An Overview of Gush Jeannie Albrecht David Irwin
The Who, What, Why and How of High Performance Computing Applications in the Cloud Soheila Abrishami 1.
Project Overview Goal: Instrumentation and Measurement capabilities for GENI experimenters and operations Outcomes: Software to perform centralized and.
Cloud Computing to Satisfy Peak Capacity Needs Case Study.
Cloud Testing – Guidelines and Approach. Agenda Understanding “The Cloud”? Why move to Cloud? Testing Philosophy Challenges Guidelines to select a Cloud.
SLA Basics Describes a set of non functional requirements of the service. Example : RTO time – Return to Operation Time if case of failure SLO – Service.
Tunis, Tunisia, 28 April 2014 Business Values of Virtualization Mounir Ferjani, Senior Product Manager, Huawei Technologies 2.
1 In VINI Veritas: Realistic and Controlled Network Experimentation Jennifer Rexford with Andy Bavier, Nick Feamster, Mark Huang, and Larry Peterson
FI-WARE – Future Internet Core Platform FI-WARE Cloud Hosting July 2011 High-level description.
1© Copyright 2015 EMC Corporation. All rights reserved. SDN INTELLIGENT NETWORKING IMPLICATIONS FOR END-TO-END INTERNETWORKING Simone Mangiante Senior.
VPN Extension Requirements for Private Clouds draft-so-vepc-00.txt.
SPRING 2011 CLOUD COMPUTING Cloud Computing San José State University Computer Architecture (CS 147) Professor Sin-Min Lee Presentation by Vladimir Serdyukov.
Bandwidth Measurements for VMs in Cloud Amit Gupta and Rohit Ranchal Ref. Cloud Monitoring Framework by H. Khandelwal, R. Kompella and R. Ramasubramanian.
Leveraging OpenFlow for Resource Placement of Virtual Desktops Project Team: Prasad Calyam, Ph.D. Sudharsan Rajagopalan,
Virtual Desktop Infrastructure Solution Stack Cam Merrett – Demonstrator User device Connection Bandwidth Virtualisation Hardware Centralised desktops.
VAP What is a Virtual Application ? A virtual application is an application that has been optimized to run on virtual infrastructure. The application software.
Component 4: Introduction to Information and Computer Science Unit 10: Future of Computing Lecture 2 This material was developed by Oregon Health & Science.
COnvergence of fixed and Mobile BrOadband access/aggregation networks Work programme topic: ICT Future Networks Type of project: Large scale integrating.
Naixue GSU Slide 1 ICVCI’09 Oct. 22, 2009 A Multi-Cloud Computing Scheme for Sharing Computing Resources to Satisfy Local Cloud User Requirements.
Cloud computing is the use of computing resources (hardware and software) that are delivered as a service over the Internet. Cloud is the metaphor for.
Cloud Testing Speaker : Mrityunjaya Hikkalgutti Date : 3 rd July 2010.
Hybrid Cloud Experiments with GENI for Multi-site Opt-in Use Cases Prasad Calyam (PI) Collaborators: Ray Leto (TotalSim), Rob Kopp (Metro Data Center)
OnTimeMeasure Integration with Gush Prasad Calyam, Ph.D. (PI) Tony Zhu (Software Programmer) Alex Berryman (REU Student) GEC10 Selected.
Distributed Real-Time Systems for the Intelligent Power Grid Prof. Vincenzo Liberatore.
NICE :Network Intrusion Detection and Countermeasure Selection in Virtual Network Systems.
Virtualization. Virtualization  In computing, virtualization is a broad term that refers to the abstraction of computer resources  It is "a technique.
Sponsored by the National Science Foundation Research & Experiments on GENI GENI CC-NIE Workshop NSF Mark Berman, Mike Zink January 7,
Network Aware Resource Allocation in Distributed Clouds.
Component 4: Introduction to Information and Computer Science Unit 10b: Future of Computing.
Cluster Reliability Project ISIS Vanderbilt University.
Software-defined Networking Capabilities, Needs in GENI for VMLab ( Prasad Calyam; Sudharsan Rajagopalan;
1 Supporting the development of distributed systems CS606, Xiaoyan Hong University of Alabama.
Autonomic SLA-driven Provisioning for Cloud Applications Nicolas Bonvin, Thanasis Papaioannou, Karl Aberer Presented by Ismail Alan.
Network Plus Virtualization Concepts. Virtualization Overview Virtualization is the emulation of a computer environment called a Virtual Machine. A Hypervisor.
GridStat on GENI: Simulating a Smart Power Grid Infrastructure over GENI Divya Giri, Ruma Paul, Haiqin Liu, Victor Valgenti, Carl Hauser and Min Sik Kim.
GENI Experiments in Optimizing Network Environments using XSP Ezra Kissel and Martin Swany University of Delaware Abstract Our proposal is to build, deploy.
Sponsored by the National Science Foundation GENI Exploring Networks of the Future
Secure Opportunistic Mobile Application Offload for Enterprise Networks Aaron Gember and Aditya Akella University of Wisconsin – Madison Abstract Application-independent.
OnTimeMeasure-GENI: Centralized and Distributed Measurement Orchestration Software Prasad Calyam, Ph.D. (PI) Paul Schopis, (Co-PI) Weiping Mandrawa (Network.
The Performance Evaluation of Intra-domain Bandwidth Allocation and Inter-domain Routing Algorithms for a QoS-guaranteed Routing Path Discovery Bo Li,
IBM Bluemix Ecosystem Development Hands on Workshop Section 1 - Overview.
 The End to the Means › (According to IBM ) › 03.ibm.com/innovation/us/thesmartercity/in dex_flash.html?cmp=blank&cm=v&csr=chap ter_edu&cr=youtube&ct=usbrv111&cn=agus.
Paperless Timesheet Management Project Anant Pednekar.
Copyright © 2005 VMware, Inc. All rights reserved. How virtualization can enable your business Richard Allen, IBM Alliance, VMware
Globus and PlanetLab Resource Management Solutions Compared M. Ripeanu, M. Bowman, J. Chase, I. Foster, M. Milenkovic Presented by Dionysis Logothetis.
Copyright © 2010, Performance and Power Management for Cloud Infrastructures Hien Nguyen Van; Tran, F.D.; Menaud, J.-M. Cloud Computing (CLOUD),
Web Technologies Lecture 13 Introduction to cloud computing.
Use-cases for GENI Instrumentation and Measurement Architecture Design Prasad Calyam, Ph.D. (PI – OnTimeMeasure, Project #1764) March 31.
Capacity Planning in a Virtual Environment Chris Chesley, Sr. Systems Engineer
Office of Administration Enterprise Server Farm September 2008 Briefing.
Architecture for Resource Allocation Services Supporting Interactive Remote Desktop Sessions in Utility Grids Vanish Talwar, HP Labs Bikash Agarwalla,
Resource Optimization for Publisher/Subscriber-based Avionics Systems Institute for Software Integrated Systems Vanderbilt University Nashville, Tennessee.
 Cloud Computing technology basics Platform Evolution Advantages  Microsoft Windows Azure technology basics Windows Azure – A Lap around the platform.
Building on virtualization capabilities for ExTENCI Carol Song and Preston Smith Rosen Center for Advanced Computing Purdue University ExTENCI Kickoff.
Fermilab Scientific Computing Division Fermi National Accelerator Laboratory, Batavia, Illinois, USA. Off-the-Shelf Hardware and Software DAQ Performance.
Chapter 1: Introduction
Architecture and Algorithms for an IEEE 802
LIGHTWEIGHT CLOUD COMPUTING FOR FAULT-TOLERANT DATA STORAGE MANAGEMENT
Software Architecture in Practice
2016 Citrix presentation.
Introduction to Cloud Computing
Conditions leading to the rise of virtual machines
Managing Clouds with VMM
GENI Exploring Networks of the Future
Presentation transcript:

Resource Allocation in Virtual Desktop Clouds: VMLab-GENI Experiment Rohit Patali, Prasad Calyam, Mukundan Sridharan, Alex Berryman The Ohio State University, Columbus Abstract User communities are rapidly transitioning their "traditional desktops" that have dedicated hardware and software installations into "virtual desktop clouds" (VDCs) that are accessible via thin-clients. To allocate and manage VDC resources for Internet-scale desktop delivery, existing works focus mainly on managing server-side resources based on utility functions of CPU and memory loads, and do not consider network health and thin-client user experience. We present an analytical model viz., "Utility-Directed Resource Allocation Model (U-RAM)" to solve the resource allocation problem within VDCs. Our solution leverages utility functions of system, network and human components obtained using a novel virtual desktop performance benchmarking toolkit viz., "VDBench" that we developed. We evaluate our solution on GENI with varying user load and network health conditions. Evaluation results demonstrate that our solution maximizes VDC scalability i.e., 'VDs per core density', and 'user connections quantity', while delivering satisfactory thin-client user experience. Research Objectives Develop “system-aware”, “network-aware”, “human-aware” frameworks and tools to deploy virtual desktop clouds Couple client-and-server resource adaptation with measurements of network health and user experience to: - Minimize cloud resource over-commitment - Avoid guesswork in configuring thin client protocols - Deliver optimum user experience of virtual applications VDCs Today – Guesswork and Overprovisioning VDC Resource Allocation Scheme Current and Proposed Publications U-RAM Illustration Planned GENI Demo Cloud Scalability Performance Comparision The Research efforts have resulted in the following publications: Conference / Journal Papers Alex Berryman, Prasad Calyam, Albert Lai, Matthew Honigford, "VDBench: A Benchmarking Toolkit for Thin- client based Virtual Desktop Environments", IEEE Conference on Cloud Computing Technology and Science (CloudCom), Under Work: “Utility-Directed Resource Allocation in Virtual Desktop Clouds” The Research Effort hopes to result in the following theses/dissertations: Master’s Thesis: Resource Placement and Provisioning in Virtual Desktop Cloud – Rohit Patali Use of Glab/GENI Infrastructure Extend VMLab to a virtual desktop cloud with 3 data centers using the NSF GENI Testbed Facility Resource nodes in ProtoGENI/PlanetLab e.g., ~30 MHz CPU and ~15 GB RAM to install VMware ESX and support ~15 VD users VMLab data center will be identical to the ProtoGENI/PlanetLab setup Rate limit all VD load at data centers to 10 Mbps network bandwidth using a network emulator OnTimeMeasure Node Beacons will be installed at all the thin-clients and data centers; Root Beacon will be installed at VMLab Gush tool will be used from demo site to: - instruct VMLab web-portal to send load control commands to the smart thin-clients in ProtoGENI PlanetLab - Control OnTimeMeasure measurement service Future Work Provision “sandbox” and “desktop” VMs within a slice - For GENI experimenters, Classroom Labs, Internet users, etc. - VMs will host trial and open-source software for users Users generate synchronous and asynchronous loads Profile “monitor” and “user” VMs under various load conditions and investigate decision schemes for resource allocation - Monitor” VMs are instrumented to perform experiment runs - User loads trigger performance data logging in monitor VMs Experiments A utility function indicates how much of application performance can be increased with larger resource allocation. Beyond a certain point, application utility saturates and any additional resource allocation fails to further increase application performance. Fixed RAM (F-RAM) tends to allocate resources that result in Q excess U-RAM profiles users based on VDBench measurements and allocates resources that results in either Q min /Q set /Q max Satisfies SLA along timeliness and coding efficiency quality dimensions and ensures optimum user experience based on resources available. 1 st DFG/GENI Doctoral Consortium, San Juan, PR March 13 th -15 th, 2011 Home User Mobile User VDC Service Provider Inadequate CPU, memory and bandwidth (Impact e.g., Slow interaction response times) Calls from unhappy customers High operation $$ Problem: Resource allocation without awareness of system, network and user experience characteristics Inadequate CPU, memory and bandwidth (Impact e.g., IPTV with impairments and slow playback) Excess CPU, memory and bandwidth (Impact e.g., Good interaction response times and smooth IPTV playback) Research Scientist CPU Memory Bandwidth I. New VD requests handling with freely available resources II. New VD requests handling with all available resources allocated III. New VD request rejected when SLA violation situation occurs Legend: