Datacenter Traffic Measurement and Classification

Similar documents
Priority-aware Coflow Placement and scheduling in Datacenters

On the road again. The network layer. Data and control planes. Router forwarding tables. The network layer data plane. CS242 Computer Networks

Transmission Control Protocol Introduction

Link-layer switches. Jurassic Park* LANs with backbone hubs are good. LANs with backbone hubs are bad. Hubs, bridges, and switches

Due Date: Lab report is due on Mar 6 (PRA 01) or Mar 7 (PRA 02)

Cisco Tetration Analytics, Release , Release Notes

(ii). o IP datagram packet is payload of a TCP segment o TCP segment is payload of an IP datagram. (iii).

Chapter 4. IP Addresses: Classful Addressing. PDF created with FinePrint pdffactory Pro trial version

2. When an EIGRP-enabled router uses a password to accept routes from other EIGRP-enabled routers, which mechanism is used?

Admin Report Kit for Exchange Server

FIREWALL RULE SET OPTIMIZATION

2. When logging is used, which severity level indicates that a device is unusable?

Chapter 6 Delivery and Routing of IP Packets. PDF created with FinePrint pdffactory Pro trial version

Telecommunication Protocols Laboratory Course

CCNA 3 Chapter 2 v5.0 Exam Answers 2015 (100%)

CCNA 1 Chapter v5.1 Answers 100%

The programming for this lab is done in Java and requires the use of Java datagrams.

IRDS: Data Mining Process

A Characterization of Data Mining Algorithms on a Modern Processor

IT Essentials (ITE v6.0) Chapter 7 Exam Answers 100% 2016

Please contact technical support if you have questions about the directory that your organization uses for user management.

Machine-Learning-Based Flow scheduling in OTSSenabled

CCNA 1 Chapter v5.1 Answers 100%

App Orchestration 2.6

Usage of Cooperative Communication in Mobile Adhoc Networks

PCI Multiple I/O Host Adapter 1. Introduction

Linking network nodes

CCNA 1 Chapter v5.1 Answers 100%

Cisco EPN Manager Network Administration

Retrieval Effectiveness Measures. Overview

EView/400i Management Pack for Systems Center Operations Manager (SCOM)

CCNA 1 Chapter v5.1 Answers 100%

CCNA 1 Chapter 1 v5.03 Exam Answers 2016

SafeDispatch SDR Gateway for MOTOROLA TETRA

This labs uses traffic traces from Lab 1 and traffic generator and sink components from Lab 2.

CMU 15-7/381 CSPs. Teachers: Ariel Procaccia Emma Brunskill (THIS TIME) With thanks to Ariel Procaccia and other prior instructions for slides

Chapter 2. The OSI Model and TCP/IP Protocol Suite. PDF created with FinePrint pdffactory Pro trial version

An Introduction to Crescendo s Maestro Application Delivery Platform

Cisco EPN Manager Operations

WLAN Interoperability Testing with Expert Analysis

CS4500/5500 Operating Systems Synchronization

Troubleshooting of network problems is find and solve with the help of hardware and software is called troubleshooting tools.

Software Defined Networking and OpenFlow. Jeffrey Dalla Tezza and Nate Schloss

Cisco EPN Manager Network Administration - Optical

The transport layer. Transport-layer services. Transport layer runs on top of network layer. In other words,

Elasticity : Advanced Cloud Computing. Garth Gibson Greg Ganger Majd Sakr. Feb 6, Adv. Cloud Computing 1

EcoStruxure for Data Centers FAQ

Moving packets. Moving datagrams. Suppose host A want to send IP to host B. Host A wants to send to host E. Generalized forwarding and SDN

Data Communications over Context-Based WMNs Delay Performance Evaluation

Questions and Answers

Experience With Processes and Monitors in Mesa

Speculative Parallelization. Devarshi Ghoshal

HP OpenView Performance Insight Report Pack for Quality Assurance

Integrated Netflow and IP Routing Analysis

HP Server Virtualization Solution Planning & Design

Using SPLAY Tree s for state-full packet classification

CCNA 3 Chapter 8 v5.0 Exam Answers 2015 (100%) CCNA 5 Page 1

Vulnerability Protection A Buffer for Patching

LIN101 RS232 / LAN INTERFACE

Iowa State University

Hierarchical Classification of Amazon Products

2. What is the most cost-effective method of solving interface congestion that is caused by a high level of traffic between two switches?

This document lists hardware and software requirements for Connected Backup

HPE AppPulse Mobile. Software Version: 2.1. IT Operations Management Integration Guide

These tasks can now be performed by a special program called FTP clients.

COSC 4397 Parallel Computation. Performance Modeling. Edgar Gabriel. Spring Motivation

1. What are two disadvantages of employing teleworkers in an organization? (Choose two.)

High Speed Shared Memory Networks in Hardware in the Loop Applications

Networks: Communicating and Sharing Resources. Chapter 7: Networks: Communicating and Sharing Resources

UFuRT: A Work-Centered Framework and Process for Design and Evaluation of Information Systems

Chapter 5. The Network Layer IP

CCNA Security v2.0 Chapter 10 Exam Answers

Data Link Layer 10/28/2013. Data Link Layer. Adaptors Communicating. Data Link Layer. Session Transport Network Data Link Physical

1. What is a characteristic of Frame Relay that provides more flexibility than a dedicated line?

Think of transport technology to support ultra-high bandwidth and/or ultra-low latency draft-han-6man-in-band-signaling-for-transport-qos

ELEC5509 Mobile Networks

Contact Center. Service Description for Release 2016 R4. April, 2017

Automated Canopy Estimator(ACE): Enhancing Crop Modelling and Decision Making in Agriculture. A. D. Coy, D.R. Rankine, M.A. Taylor, and D. C.

Dynamically Provisioned Networks as a Substrate for Science. David Foster CERN

H A N D O U T : W I N D O W S N E T W O R K I N G

Accession Communicator for Desktop User Guide Hosted IP Phone System

CSE 3320 Operating Systems Computer and Operating Systems Overview Jia Rao

SSDNow vs. HDD and Use Cases/Scenarios. U.S.T.S. Tech. Comm

INVENTION DISCLOSURE

ACCELERATOR 7.0 CISCO UNIFIED COMMUNICATION MANAGER 10.5 INTEGRATION GUIDE

CCNA 1 v5.1 Practice Final Exam Answers %

Memory Hierarchy. Goal of a memory hierarchy. Typical numbers. Processor-Memory Performance Gap. Principle of locality. Caches

ERAC. Efficient and Robust Architecture for the big data Cloud. Tor Skeie

PL-2302 Mac OS Driver MAC/PC and PC/MAC Communication Software

Hands-on Windows Azure Application Architecture & Development (3 days)

Case Metrics Guide. January 11, 2019 Version For the most recent version of this document, visit our documentation website.

Student participation Students can register online, track progress, express interest and demonstrate proficiency.

Operational Security. Speaking Frankly The Internet is not a very safe place. A sense of false security... Firewalls*

Comp 245 Data Structures. Queues

Packet erasure coding with random access to reduce losses of delay sensitive multislot messages

CSE 3320 Operating Systems Page Replacement Algorithms and Segmentation Jia Rao

Chapter 10: Information System Controls for System Reliability Part 3: Processing Integrity and Availability

1 Version Spaces. CS 478 Homework 1 SOLUTION

CSE 3320 Operating Systems Deadlock Jia Rao

Australian Statistics API Specification

Transcription:

Datacenter Traffic Measurement and Classificatin Speaker: Lin Wang Research Advisr: Biswanath Mukherjee

Grup meeting 6/15/2017 Datacenter Traffic Measurement and Analysis Data Cllectin Cllect netwrk events frm each f 1500 servers Fr ver tw mnths. Traffic Characteristics Server pairs within the same rack mre likely t exchange mre bytes. 21% prbability t exchange data within the same rack. 0.5% prbability t exchange data in different racks. Kandula S, Sengupta S, Greenberg A, et al. The nature f data center traffic: measurements & analysis[c]//prceedings f the 9th ACM SIGCOMM cnference n Internet measurement cnference. ACM, 2009: 202-208.

Grup meeting 6/15/2017 Datacenter Traffic Measurement and Analysis Traffic Characteristics Server either talks t almst all the ther servers within the rack (the bump near 1 in figure left) r fewer than 25% f servers within the rack. Server either desn t talk t servers utside its rack (the spike at zer in figure right) r it talks t abut 1-10% f utside servers.

Grup meeting 6/15/2017 Datacenter Traffic Measurement and Analysis Traffic Characteristics Cmpare the rates f flws that verlap high utilizatin perids. Rates d nt change appreciably (see cdf belw). Errrs(e.g. flw timeuts r failure) is nt visible in flw rates. Hence we crrelate high utilizatin epchs directly with applicatin level lgs.

Grup meeting 6/15/2017 Datacenter Traffic Measurement and Analysis Traffic Characteristics Traffic mix changes frequently. The figure plts the duratins f millin flws (a day s wrth f flws) in the cluster. Mst flws cme and g (80% last less than 10s) and there are few lng running flws (less than 0.1% last lnger than 200s).

Grup meeting 6/15/2017 Datacenter Traffic Classificatin Why d we need t identify elephant flw? Previus paper shws that a large fractin f datacenter traffic is carried in a small fractin f flws. 90% f the flws carry less than 1MB f data >90% f bytes transferred are in flws greater than 100MB. Hash-based flw frwarding techniques (e.g. Equal-Cst Multi-Path (ECMP) ruting) wrks well nly fr mice flws and n elephant flws.

Grup meeting 6/15/2017 Mice flw VS elephant flw Small size packet Shrt flw Large number Shrt-lived Large size packet Large vlume flw Small number Lng lasting

Grup meeting 6/15/2017 Mice flw VS elephant flw If we nly care abut the number f packets in the queue, elephant flw transmissin is easy t be degraded. If we nly care abut the ttal size f packets in the queue, mice flw transmissin is easy t be degraded. < >

Grup meeting 6/15/2017 Datacenter Traffic Classificatin Slutin: Mahut, a lw-verhead yet effective traffic management system. End-hst-based elephant detectin. Advantages f detecting elephant flw at end hst. Netwrk behavir f a flw is affected by hw rapidly the end-pint applicatins are generating data, and this is nt biased by cngestin in the netwrk. In cntrast t in-netwrk mnitrs, the end hst OS has better visibility int the behavir f applicatins. In datacenters, it is pssible t augment the end hst OS; this is enabled by the single administrative dmain and sftware unifrmity typical f mdern datacenters. Use very little verhead. In cntrast, using an in-netwrk mechanism t mnitr is infeasible, even n an edge switch, and even mre s n a cre switch. Curtis A R, Kim W, Yalagandula P. Mahut: Lw-verhead datacenter traffic management using end-hst-based elephant detectin[c]//infocom, 2011 Prceedings IEEE. IEEE, 2011: 1629-1637.

Grup meeting 6/15/2017 Datacenter Traffic Classificatin Mahut algrithm: Usea shim layer in the end hsts t mnitr the scket buffers.

Grup meeting 6/15/2017 Datacenter Traffic Classificatin Simulatin parameters:

Grup meeting 6/15/2017 Datacenter Traffic Classificatin Simulatin results:

Grup meeting 6/15/2017 Machine Learning fr Traffic Flw Classificatin Machine Learning Algrithms: Naïve-Bayes (NBD, NBK) C4.5 Decisin Tree Bayesian Netwrk Naïve Bayes Tree Flw and Feature Definitins Limitatin: Packet paylad independent Transprt layer independent Cntext limited t a single flw (i.e. n feature spanning multiple flws) Simple t cmpute Williams N, Zander S, Armitage G. A preliminary perfrmance cmparisn f five machine learning algrithms fr practical IP traffic flw classificatin[j]. ACM SIGCOMM Cmputer Cmmunicatin Review, 2006, 36(5): 5-16.

Grup meeting 6/15/2017 Machine Learning fr Traffic Flw Classificatin Feature Candidates Prtcl Flw duratin Flw vlume in bytes and packets Packet length (minimum, mean, maximum and standard deviatin) Inter-arrival time between packets (minimum, mean, maximum and standard deviatin). Feature Reductin Use CFS and CON t chse features:

Grup meeting 6/15/2017 Machine Learning fr Traffic Flw Classificatin Simulatin results

Grup meeting 6/15/2017 Machine Learning fr Traffic Flw Classificatin Simulatin results

Grup meeting 6/15/2017 amlwang@ucdavis.edu