Flexible Performance Prediction of Data Center Networks using Automatically Generated Simulation Models Piotr Rygielski, Samuel Kounev, Phuoc Tran-Gia.

Slides:



Advertisements
Similar presentations
QoS Strategy in DiffServ aware MPLS environment Teerapat Sanguankotchakorn, D.Eng. Telecommunications Program, School of Advanced Technologies Asian Institute.
Advertisements

Models and techniques for verification of Software Defined Networks
SKELETON BASED PERFORMANCE PREDICTION ON SHARED NETWORKS Sukhdeep Sodhi Microsoft Corp Jaspal Subhlok University of Houston.
The Network Weather Service A Distributed Resource Performance Forecasting Service for Metacomputing Rich Wolski, Neil T. Spring and Jim Hayes Presented.
Katz, Stoica F04 EECS 122 Introduction to Computer Networks (Fall 2003) Network simulator 2 (ns-2) Department of Electrical Engineering and Computer Sciences.
Internet Traffic Patterns Learning outcomes –Be aware of how information is transmitted on the Internet –Understand the concept of Internet traffic –Identify.
NoC Modeling Networks-on-Chips seminar May, 2008 Anton Lavro.
A BitTorrent Module for the OMNeT++ Simulator MASCOTS 2009, London, UK G. Xylomenos (with K. Katsaros, V.P. Kemerlis and C. Stais)
A Model-Driven Framework for Architectural Evaluation of Mobile Software Systems George Edwards Dr. Nenad Medvidovic Center.
1 Distributed Online Simultaneous Fault Detection for Multiple Sensors Ram Rajagopal, Xuanlong Nguyen, Sinem Ergen, Pravin Varaiya EECS, University of.
Glenn Research Center at Lewis Field Deep Space Network Emulation Shaun Endres and Behnam Malakooti Case Western Reserve University Department of Electrical.
Hardware & Software Needed For LAN and WAN
Copyright © 2012, QoS-aware Network Operating System for Software Defined Networking with Generalized OpenFlows Kwangtae Jeong, Jinwook Kim.
Layer-3 Routing Natawut Nupairoj, Ph.D. Department of Computer Engineering Chulalongkorn University.
Not All Microseconds are Equal: Fine-Grained Per-Flow Measurements with Reference Latency Interpolation Myungjin Lee †, Nick Duffield‡, Ramana Rao Kompella†
Switching, routing, and flow control in interconnection networks.
Process-oriented System Automation Executable Process Modeling & Process Automation.
Włodzimierz Funika, Filip Szura Automation of decision making for monitoring systems.
FPGA for Underwater Communication Pradyumna (Prad) Kadambi Mentor: Cody Youngbull April 13, 2015.
OMNET++. Outline Introduction Overview The NED Language Simple Modules.
Document Number ETH West Diamond Avenue - Third Floor, Gaithersburg, MD Phone: (301) Fax: (301)
Composing Software Defined Networks Jennifer Rexford Princeton University With Joshua Reich, Chris Monsanto, Nate Foster, and.
(1) Univ. of Rome Tor Vergata, (2) Consortium GARR, (3) CREATE-NET
Redes Inalámbricas Máster Ingeniería de Computadores 2008/2009 Tema 7.- CASTADIVA PROJECT Performance Evaluation of a MANET architecture.
Sidewinder A Predictive Data Forwarding Protocol for Mobile Wireless Sensor Networks Matt Keally 1, Gang Zhou 1, Guoliang Xing 2 1 College of William and.
Visualisation and Analysis of Real Time Application Behaviour in a Simulated Network (!Temporal Databases  K. Maciunas) Evan Bourlotos Supervisors Cheryl.
1 Enabling Large Scale Network Simulation with 100 Million Nodes using Grid Infrastructure Hiroyuki Ohsaki Graduate School of Information Sci. & Tech.
Yuan Xue Vanderbilt University
VeriFlow: Verifying Network-Wide Invariants in Real Time
Software Framework for Teleoperated Vehicles Team Eye-Create ECE 4007 L01 Karishma Jiva Ali Benquassmi Safayet Ahmed Armaghan Mahmud Khin Lay Nwe.
Application Redundancy Tool A.R.T. CS 495 Fall 2005 Kristi Olson.
Imperial College - Department of Computing Continuous Performance Testing in Virtual Time Nikos Baltas & Tony Field Department of Computing Imperial College.
SpaceWire Plug-and-Play: A Roadmap Peter Mendham, Albert Ferrer Florit, Steve Parkes Space Technology Centre, University of Dundee 1.
11 Experimental and Analytical Evaluation of Available Bandwidth Estimation Tools Cesar D. Guerrero and Miguel A. Labrador Department of Computer Science.
OBJECTIVE: o Describe various network topologies o Discuss the role of network devices o Understand Network Configuration Factors to deploy a new network.
Software Defined Networks for Dynamic Datacenter and Cloud Environments.
PRoPHET+: An Adaptive PRoPHET- Based Routing Protocol for Opportunistic Network Ting-Kai Huang, Chia-Keng Lee and Ling-Jyh Chen.
NetOpen Networking Service: Software-defined Networking Service on Programmable Network Substrates Namgon Kim and JongWon Kim Networked Computing Systems.
Functional Verification of Dynamically Reconfigurable Systems Mr. Lingkan (George) Gong, Dr. Oliver Diessel The University of New South Wales, Australia.
Trajectory Sampling for Direct Traffic Oberservation N.G. Duffield and Matthias Grossglauser IEEE/ACM Transactions on Networking, Vol. 9, No. 3 June 2001.
Design, Implementation and Tracing of Dynamic Backpressure Routing for ns-3 José Núñez-Martínez Research Engineer Centre Tecnològic de Telecomunicacions.
An Efficient Gigabit Ethernet Switch Model for Large-Scale Simulation Dong (Kevin) Jin.
1 RealProct: Reliable Protocol Conformance Testing with Real Nodes for Wireless Sensor Networks Junjie Xiong, Edith C.-Ngai, Yangfan Zhou, Michael R. Lyu.
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 1 Based upon slides from Jay Lepreau, Utah Emulab Introduction Shiv Kalyanaraman
Modeling Virtualized Environments in Simalytic ® Models by Computing Missing Service Demand Parameters CMG2009 Paper 9103, December 11, 2009 Dr. Tim R.
An Efficient Gigabit Ethernet Switch Model for Large-Scale Simulation Dong (Kevin) Jin.
Computer Simulation of Networks ECE/CSC 777: Telecommunications Network Design Fall, 2013, Rudra Dutta.
Efficient Gigabit Ethernet Switch Models for Large-Scale Simulation Dong (Kevin) Jin David Nicol Matthew Caesar University of Illinois.
An Efficient Gigabit Ethernet Switch Model for Large-Scale Simulation Dong (Kevin) Jin.
An Efficient Gigabit Ethernet Switch Model for Large-Scale Simulation Dong (Kevin) Jin.
for SDN-based flow handover in wireless environments Daniel Corujo Carlos Guimarães Rui L. Aguiar
Research Unit for Integrated Sensor Systems and Oregano Systems Cern Timing Workshop 2008 Patrick Loschmidt, Georg Gaderer, and Nikolaus Kerö.
Simulation of O2 offline processing – 02/2015 Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture Eugen Mudnić.
CS 283Computer Networks Spring 2013 Instructor: Yuan Xue.
@Yuan Xue CS 283Computer Networks Spring 2011 Instructor: Yuan Xue.
Routing Semester 2, Chapter 11. Routing Routing Basics Distance Vector Routing Link-State Routing Comparisons of Routing Protocols.
Deterlab Tutorial CS 285 Network Security. What is Deterlab? Deterlab is a security-enhanced experimental infrastructure (based on Emulab) that supports.
Traffic Simulation L3b – Steps in designing a model Ing. Ondřej Přibyl, Ph.D.
Software Defined Networking BY RAVI NAMBOORI. Overview  Origins of SDN.  What is SDN ?  Original Definition of SDN.  What = Why We need SDN ?  Conclusion.
Realistic Mobility Models for Vehicular Ad hoc Network (VANET) Simulations ITST 高弘毅 洪佳瑜 蔣克欽.
SDN challenges Deployment challenges
FlowRadar: A Better NetFlow For Data Centers
Taming the Complexity of Artifact Reproducibility
Author: Daniel Guija Alcaraz
Computer Simulation of Networks
Peter Poplavko, Saddek Bensalem, Marius Bozga
Providing QoS through Active Domain Management
Modeling of Parametric Dependencies for Performance Prediction of Component-based Software Systems at Run-time Simon Eismann, Jürgen Walter, Joakim Kistowski,
Configuration DB Status report Lana Abadie
In-network computation
Presentation transcript:

Flexible Performance Prediction of Data Center Networks using Automatically Generated Simulation Models Piotr Rygielski, Samuel Kounev, Phuoc Tran-Gia Chair of Software Engineering University of Würzburg SIMUtools2015, Athens, Greece,

Motivation 2 Piotr Rygielski (dst_IP>*.*.*.128) ? port1 : port0; (src_TCP==80 && src_TCP==443) ? port1 : port0; What if… Current performance known – monitoring. Goal: predict performance after a change.

Research Gap 3 Piotr Rygielski  End-to-end performance analysis not detailed enough  Existing network models too coarse or too fine grained  Other approaches focus only on selected technologies/protocols  Flexibility in modeling is missing Black-box modelsDetailed simulations Time overhead Accuracy Model

Approach 4 Piotr Rygielski Real network Model extraction Model transformation(s) Descriptive model Performance model(s)

Approach 5 Piotr Rygielski Real network DNI meta model (modeling language) Structure model Traffic model Configuration model Model-to-model transformations to QN to OMNeT++ to QPN to ns3 to formulas other... Performance models single model script

Models and Transformations 6 Piotr Rygielski

miniDNI Meta-Model 7 Piotr Rygielski  When not enough data to build full DNI instance  Very coarse-granular modeling

DNI Meta-Model Structure model Traffic model Configuration model SoftwareComponent NetworkInterface Link PerformanceDescriptions Node TrafficSource Workload Flow Start Stop Wait Transmit Loop Sequence Route ProtocolStack NetworkProtocol DNI Meta-Model (short) 8 Piotr Rygielski

Transformation mDNI-to-QPN 9 Piotr Rygielski  QPN model of a network node, e.g., Switch, Server (mDNI)  Aspects: None, Generator, Receiver, Traversal

Transformation mDNI-to-QPN 10 Piotr Rygielski  QPN model of a network link (mDNI)  Delays from Interfaces and links integrated in queueing place

Transformation mDNI-to-QPN 11 Piotr Rygielski

Transformations - comparison 12 Piotr Rygielski

Traffic Management System GPS Sensors Traffic Light Sensors Induction Loops Traffic Cameras Case study – SBUS/PIRATES 13 Piotr Rygielski

Case study – SBUS/PIRATES 14 Piotr Rygielski

Case study – SBUS/PIRATES 15 Piotr Rygielski

Model Calibration 16 Piotr Rygielski

Experiment - Hardware 17 Piotr Rygielski

Results – Prediction Accuracy 18 Piotr Rygielski Motivation & ApproachDNI & TransformationsCurrent FocusPlanning

Results – Simulation Time 19 Piotr Rygielski Motivation & ApproachDNI & TransformationsCurrent FocusPlanning  Dumbbell topology

Results – Simulation Time 20 Piotr Rygielski Motivation & ApproachDNI & TransformationsCurrent FocusPlanning  Dumbbell topology

Conclusions 21 Piotr Rygielski Motivation & ApproachDNI & TransformationsCurrent FocusPlanning Automatically generated three predictive modelsPrediction errors up to 18% for DNI (fully automatic process)miniDNI-QPN: accuracy loss (~4%) with speedup up to 300xSupport for network virtualization in DNI (SDN planned)Model calibration is important. Modeling support tools needed

Thank You! Code & more info: