Bandwidth Measurements for VMs in Cloud Amit Gupta and Rohit Ranchal Ref. Cloud Monitoring Framework by H. Khandelwal, R. Kompella and R. Ramasubramanian.

Slides:



Advertisements
Similar presentations
Network Resource Broker for IPTV in Cloud Computing Lei Liang, Dan He University of Surrey, UK OGF 27, G2C Workshop 15 Oct 2009 Banff,
Advertisements

Cloud Service Models and Performance Ang Li 09/13/2010.
Towards Predictable Datacenter Networks
SLA-Oriented Resource Provisioning for Cloud Computing
Ningning HuCarnegie Mellon University1 Optimizing Network Performance In Replicated Hosting Peter Steenkiste (CMU) with Ningning Hu (CMU), Oliver Spatscheck.
LOAD BALANCING IN A CENTRALIZED DISTRIBUTED SYSTEM BY ANILA JAGANNATHAM ELENA HARRIS.
1 NETE4631 Cloud deployment models and migration Lecture Notes #4.
An Approach to Secure Cloud Computing Architectures By Y. Serge Joseph FAU security Group February 24th, 2011.
Transparent Checkpoint of Closed Distributed Systems in Emulab Anton Burtsev, Prashanth Radhakrishnan, Mike Hibler, and Jay Lepreau University of Utah,
Look Who’s Talking: Discovering Dependencies between Virtual Machines Using CPU Utilization HotCloud 10 Presented by Xin.
Dynamically Scaling Applications in the Cloud Presented by Paul.
Scalable and Crash-Tolerant Load Balancing based on Switch Migration
COMMA: Coordinating the Migration of Multi-tier applications 1 Jie Zheng* T.S Eugene Ng* Kunwadee Sripanidkulchai† Zhaolei Liu* *Rice University, USA †NECTEC,
On the Geographic Distribution of On- line Game Servers and Players Wu-chang FengWu-chi Feng Presented By: Abhishek Gupta.
1 Virtual Machine Resource Monitoring and Networking of Virtual Machines Ananth I. Sundararaj Department of Computer Science Northwestern University July.
Automatic Run-time Adaptation in Virtual Execution Environments Ananth I. Sundararaj Advisor: Peter A. Dinda Prescience Lab Department of Computer Science.
Service Differentiated Peer Selection An Incentive Mechanism for Peer-to-Peer Media Streaming Ahsan Habib, Member, IEEE, and John Chuang, Member, IEEE.
Peer-to-Peer Based Multimedia Distribution Service Zhe Xiang, Qian Zhang, Wenwu Zhu, Zhensheng Zhang IEEE Transactions on Multimedia, Vol. 6, No. 2, April.
On the Geographic Distribution of On- line Game Servers and Players Wu-chang FengWu-chi Feng Discussion moderated By: John Carter.
FI-WARE – Future Internet Core Platform FI-WARE Cloud Hosting July 2011 High-level description.
Inferring the Topology and Traffic Load of Parallel Programs in a VM environment Ashish Gupta Resource Virtualization Winter Quarter Project.
ProActive Routing In Scalable Data Centers with PARIS Joint work with Dushyant Arora + and Jennifer Rexford* + Arista Networks *Princeton University Theophilus.
Scale-out File Server Cluster Hyper-V Cluster Virtual Machines SMB3 Storage Network Fabric.
Inferring the Topology and Traffic Load of Parallel Programs in a VM environment Ashish Gupta Peter Dinda Department of Computer Science Northwestern University.
Jennifer Rexford Princeton University MW 11:00am-12:20pm Data-Center Traffic Management COS 597E: Software Defined Networking.
Private Cloud or Dedicated Hosts Mason Mabardy & Matt Maples.
Installing and Setting up mongoDB replica set PREPARED BY SUDHEER KONDLA SOLUTIONS ARCHITECT.
Measuring zSeries System Performance Dr. Chu J. Jong School of Information Technology Illinois State University 06/11/2012 Sponsored in part by Deer &
A User Experience-based Cloud Service Redeployment Mechanism KANG Yu.
Middleware Enabled Data Sharing on Cloud Storage Services Jianzong Wang Peter Varman Changsheng Xie 1 Rice University Rice University HUST Presentation.
Supporting GPU Sharing in Cloud Environments with a Transparent
INTRODUCTION TO CLOUD COMPUTING CS 595 LECTURE 7 2/23/2015.
Department of Computer Science Engineering SRM University
Monitoring Latency Sensitive Enterprise Applications on the Cloud Shankar Narayanan Ashiwan Sivakumar.
Network Aware Resource Allocation in Distributed Clouds.
Creating an EC2 Provisioning Module for VCL Cameron Mann & Everett Toews.
Improving Disk Latency and Throughput with VMware Presented by Raxco Software, Inc. March 11, 2011.
Presented by: Sanketh Beerabbi University of Central Florida COP Cloud Computing.
CPS Welcome to a new licensing model in SPLA.
A Framework for Elastic Execution of Existing MPI Programs Aarthi Raveendran Tekin Bicer Gagan Agrawal 1.
A Framework for Elastic Execution of Existing MPI Programs Aarthi Raveendran Graduate Student Department Of CSE 1.
From Virtualization Management to Private Cloud with SCVMM 2012 Dan Stolts Sr. IT Pro Evangelist Microsoft Corporation
COMMA: Coordinating the Migration of Multi-tier Applications Jie Zheng, T. S. Eugene Ng, Zhaolei Liu Rice University Kunwadee Sripanidkulchai NECTEC, Thailand.
High Performance File System Service for Cloud Computing Kenji Kobayashi, Osamu Tatebe University of Tsukuba, JAPAN.
1 Finding Constant From Change: Revisiting Network Performance Aware Optimizations on IaaS Clouds Yifan Gong, Bingsheng He, Dan Li Nanyang Technological.
11 Experimental and Analytical Evaluation of Available Bandwidth Estimation Tools Cesar D. Guerrero and Miguel A. Labrador Department of Computer Science.
Introduction to dCache Zhenping (Jane) Liu ATLAS Computing Facility, Physics Department Brookhaven National Lab 09/12 – 09/13, 2005 USATLAS Tier-1 & Tier-2.
Advanced Network Architecture Research Group 2001/11/74 th Asia-Pacific Symposium on Information and Telecommunication Technologies Design and Implementation.
Biomedical Big Data Training Collaborative biobigdata.ucsd.edu BBDTC UPDATES Biomedical Big Data Training Collaborative biobigdata.ucsd.edu.
Cloud Interoperability & Standards. Scalability and Fault Tolerance Fault tolerance is the property that enables a system to continue operating properly.
1. Introduction REU 2006-Packet Loss Distributions of TCP using Web100 Zoriel M. Salado, Mentors: Dr. Miguel A. Labrador and Cesar D. Guerrero 2. Methodology.
BNL Service Challenge 3 Status Report Xin Zhao, Zhenping Liu, Wensheng Deng, Razvan Popescu, Dantong Yu and Bruce Gibbard USATLAS Computing Facility Brookhaven.
EuroSys Doctoral Workshop 2011 Resource Provisioning of Web Applications in Heterogeneous Cloud Jiang Dejun Supervisor: Guillaume Pierre
Cloud-enabled, scalable Data Avenue service to process very large, heterogeneus data Péter Kacsuk, Ákos Hajnal MTA SZTAKI Francesco Tusa, Junaid Arshad.
Emerging applications in cloud High performance computing E-Commerce Media hosting Web hosting Content delivery... –from Amazon AWS survey 1 Emulated network.
KAASHIV INFOTECH – A SOFTWARE CUM RESEARCH COMPANY IN ELECTRONICS, ELECTRICAL, CIVIL AND MECHANICAL AREAS
Trojan Express Network II Goal: Develop Next Generation research network in parallel to production network to address increasing research data transfer.
Information Initiative Center, Hokkaido University North 11, West 5, Sapporo , Japan Tel, Fax: General.
Autonomic aspects in cloud data management Alexandra Carpen-Amarie KerData.
With the recent rise in cloud computing, applications are routinely accessing and interacting with data on remote resources. As data sizes become increasingly.
INTRODUCTION TO AMAZON WEB SERVICES (EC2). AMAZON WEB SERVICES  Services  Storage (Glacier, S3)  Compute (Elastic Compute Cloud, EC2)  Databases (Redshift,
Chen Qian, Xin Li University of Kentucky
ECE544: Software Assignment 3
Replication Middleware for Cloud Based Storage Service
Bandwidth Measurements for VMs in Cloud
Brandon Hixon Jonathan Moore
AWS Cloud Computing Masaki.
Cloud Security AWS as an example.
Cloud Security AWS as an example.
Presentation transcript:

Bandwidth Measurements for VMs in Cloud Amit Gupta and Rohit Ranchal Ref. Cloud Monitoring Framework by H. Khandelwal, R. Kompella and R. Ramasubramanian. Purdue University research report, 2010.

MOTIVATION Many applications are being deployed in cloud to leverage the scalability provided by the cloud providers. Tools provided by the cloud providers do not give performance metrics from the network perspective. Network topology is not exposed to the cloud users and the applications consider all network links to be homogeneous. Metrics such as available bandwidth, latency etc. will be more useful to the cloud users.

Experimental Evaluation Set up o 19 EC2 small instances (US East) o 342 (19*18) links among VMs o Ubuntu server version Centralized Scheduler for starting Iperf clients o Predefined serialized schedule file at each VM instance. o Schedule file contains a time stamp along with the nodes that should communicate for a single reading. * Iperf - Network testing tool to measure the network throughput between end hosts.

Experimental Evaluation Iperf takes 6 seconds to measure round trip time ( bandwidth) for a single link. Each round of measurement takes around 30 minutes for finding available bandwidth for all 342 links. Experiments runs for 5 rounds Throughput matrix: Matrix containing estimated values for available bandwidth

Bandwidth Estimation Shows the CDF of link bandwidth estimation for all the rounds. Used throughput matrix having estimated 342 values. All links in clouds are not homogeneous. Only 10% of the links have available bandwidth less than 400Mbps. All links do not get same bandwidth

Bandwidth Variation Estimation Shows the CDF of link bandwidth variation across all the rounds. Bandwidth range of a link defined as the difference between the max and min value across all rounds. For most of the links, bandwidth is consistent across time. Only 20% links have variation of more than 80 Mbps.

Virtual Machine Performance Shows the available download/upload bandwidth of all machines for a single round Almost all the machines have average available bandwidth more than 400 Mbps.

Virtual Machine Performance Shows the average available download/ upload bandwidth and its range for each machine across all rounds. Almost all the machines have average download/ upload bandwidth more than 400 Mbps. Some VMs (1, 4, 7) have large available bandwidth variation.

CONCLUSIONS Focussed on available bandwidth metric between each pair of VM instances. Amazon EC2 data center is optimally utilized with ample available bandwidth for almost all VMs. Some badly performing VMs can be pointed out based on the large variation in the available upload/download bandwidth and can be replaced with new VMs.

Future Work More performance metric such as latency etc. can be considered. These performance metrics can be used to improve the performance of applications running in the cloud. These performance metric tests can be run on large EC2 instances.

References AWS Amazon EC2: Amazon CloudWatch: Iperf Data reported in Cloud Monitoring Framework by H. Khandelwal, R. Kompella and R. Ramasubramanian. Purdue University research paper, Data provided by Khandelwal to Rohit and Amit