Technology for Backbone Web Caching Peter B Danzig Network Appliance November 1998.

Slides:



Advertisements
Similar presentations
NAGIOS AND CACTI NETWORK MANAGEMENT AND MONITORING SYSTEMS.
Advertisements

Tony Doyle - University of Glasgow GridPP EDG - UK Contributions Architecture Testbed-1 Network Monitoring Certificates & Security Storage Element R-GMA.
Equality Join R X R.A=S.B S : : Relation R M PagesN Pages Relation S Pr records per page Ps records per page.
Intermediate GNVQ ICT Computer Systems Hardware is the name that is given to any part of a computer that you can actually touch. An individual piece of.
Computer Terminology … Remember: Knowledge is Power!
Virtual Memory. Why do we need VM? Program address space: 0 – 2^32 bytes –4GB of space Physical memory available –256MB or so Multiprogramming systems.
Energy-Efficient Design Some design issues in each protocol layer Design options for each layer in the protocol stack.
CS 333 Introduction to Operating Systems Class 9 - Memory Management
Some recent work on P2P content distribution Based on joint work with Yan Huang (PPLive), YP Zhou, Tom Fu, John Lui (CUHK) August 2008 Dah Ming Chiu Chinese.
Load Sharing and Balancing - Saravanan Mathialagan Masters in Computer Science Georgia State University.
Distributed Systems Early Examples. Projects NOW – a Network Of Workstations University of California, Berkely Terminated about 1997 after demonstrating.
Sizing Router Buffers How much packet buffers does a router need? C Router Source Destination 2T The current “Rule of Thumb” A router needs a buffer size:
Multiple Processor Systems Chapter Multiprocessors 8.2 Multicomputers 8.3 Distributed systems.
Operating Systems COMP 4850/CISG 5550 Page Tables TLBs Inverted Page Tables Dr. James Money.
NetCache Architecture and Deployment Peter Danzig Network Appliance, Santa Clara, CA 元智大學 系統實驗室 陳桂慧
A Measurement Based Memory Performance Evaluation of High Throughput Servers Garba Isa Yau Department of Computer Engineering King Fahd University of Petroleum.
OCR GCSE Computing © Hodder Education 2013 Slide 1 OCR GCSE Computing Chapter 2: Memory.
GCSE Information Technology Computer Systems 2 Hardware is the name that is given to any part of a computer that you can actually touch. An individual.
Kiew-Hong Chua a.k.a Francis Computer Network Presentation 12/5/00.
Authors: Haowei Yuan and Patrick Crowley Publisher: 2013 Proceedings IEEE INFOCOM Presenter: Chia-Yi Chu Date: 2013/08/14 1.
CPSC 404, Laks V.S. Lakshmanan1 External Sorting Chapter 13: Ramakrishnan & Gherke and Chapter 2.3: Garcia-Molina et al.
Boot Sequence, Internal Component & Cisco 3 Layer Model 1.
Indexing CS 400/600 – Data Structures. Indexing2 Memory and Disk  Typical memory access: 30 – 60 ns  Typical disk access: 3-9 ms  Difference: 100,000.
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 1 Based upon slides from Jay Lepreau, Utah Emulab Introduction Shiv Kalyanaraman
Panther Media for Large Scale Development Network Architecture.
1 MSRBot Web Crawler Dennis Fetterly Microsoft Research Silicon Valley Lab © Microsoft Corporation.
Disk Basics CS Introduction to Operating Systems.
Multilevel Caches Microprocessors are getting faster and including a small high speed cache on the same chip.
Gravitational N-body Simulation Major Design Goals -Efficiency -Versatility (ability to use different numerical methods) -Scalability Lesser Design Goals.
PROOF tests at BNL Sergey Panitkin, Robert Petkus, Ofer Rind BNL May 28, 2008 Ann Arbor, MI.
BMTS 242: Computer and Systems Lecture 2: Memory, and Software Yousef Alharbi Website
Connectivity layout of the N7 switch ( ) 4* 1 Gb uplinks to backbone 2 * 10 * 1 Gb interconnections between the N7 switches 10 Gbit limits 1.3.
09/13/04 CDA 6506 Network Architecture and Client/Server Computing Peer-to-Peer Computing and Content Distribution Networks by Zornitza Genova Prodanoff.
DMBS Internals I. What Should a DBMS Do? Store large amounts of data Process queries efficiently Allow multiple users to access the database concurrently.
Western Tier 2 Site at SLAC Wei Yang US ATLAS Tier 2 Workshop Harvard University August 17-18, 2006.
1 Lecture 16: Data Storage Wednesday, November 6, 2006.
COMP SYSTEM ARCHITECTURE CACHES IN SYSTEMS Sergio Davies Feb/Mar 2014COMP25212 – Lecture 4.
1 Design and Implementation of a High-Performance Distributed Web Crawler Polytechnic University Vladislav Shkapenyuk, Torsten Suel 06/13/2006 석사 2 학기.
CS161 – Design and Architecture of Computer
Getting the Most out of Scientific Computing Resources
Determining Topology from a Capture File
Getting the Most out of Scientific Computing Resources
CS161 – Design and Architecture of Computer
Scale and Performance in the CoBlitz Large-File Distribution Service
Module 2: DriveScale architecture and components
Lecture 16: Data Storage Wednesday, November 6, 2006.
FileSystems.
Ramya Kandasamy CS 147 Section 3
Bernd Panzer-Steindel, CERN/IT
QuickPath interconnect GB/s GB/s total To I/O
Addressing: Router Design
Swapping Segmented paging allows us to have non-contiguous allocations
Architecture Background
Memory Management for Scalable Web Data Servers
CS : Technology Trends August 31, 2015 Ion Stoica and Ali Ghodsi (
COP 4600 Operating Systems Fall 2010
Jim Warner UC Santa Cruz February 2010
Distributed Content in the Network: A Backbone View
CSCI206 - Computer Organization & Programming
Data Structures and Algorithms
File Systems Implementation
Data Structures and Algorithms
CS 140 Lecture Notes: Technology and Operating Systems
CS 140 Lecture Notes: Technology and Operating Systems
CS399 New Beginnings Jonathan Walpole.
Multiple Processor Systems
Persistence: hard disk drive
Client/Server Computing and Web Technologies
ARM920T Processor This training module provides an introduction to the ARM920T processor embedded in the AT91RM9200 microcontroller.We’ll identify the.
Presentation transcript:

Technology for Backbone Web Caching Peter B Danzig Network Appliance November 1998

Speakers  Peter Danzig -- Network Appliance “Scaling Cache Deployments”  Shirish Sathaye -- Alteon Networks “Scaling w/L4-Switches”  James Aviani -- Cisco “Scaling WCCP”  Bill Maggs -- MCI/Worldcomm “Routing and Caching”  Ed Kern Digex “Report on Digex cache deployment”

Rules of thumb for Scaling Individual Cache Nodes Peter B Danzig Network Appliance November 1998

Scaling Individual Cache Nodes Rule of thumb resources required per megabit Simultaneous TCP Connections 500 Disk Needed 4 GB Disk Arms Needed 1/2-1/3 Memory 32 MB

Simultaneous Connections Little’s Law: N = W W: Average time to fetch a web page: 4-10 Seconds  URLs/Second: 25 per megabit  Number of TCP connections N = 2 * 25 * 10 = 500 connections/megabit

Disk Arms Keep disks below 50% busy A cache behind a cache sees random references Assume all references require 1 disk/ops Depending on OS, a single disk can achieve operations per second Per megabit: 25 ops/S * 2 (for head room) yields anywhere from 1 disk to 1/4 disk per megabit

Disk Space A megabit link can draw 9GB/day Peak to average load is about 4X So a 9GB disk holds the entire reference set of 4- days of operation. Assume 1/2 of references are already cached or are non-cachable…. Use 4GB of disk per megabit

Memory Each megabit means 500, dual 8 KB TCP buffers: 8 MB Proxy state for KB: 16 MB Disk buffer pool: 8MB Leads to 32MB/megabit

CPU Depends on the manufacturer and OS release For example: C720 running NetCache-3.3: 25 mbits/s C760 running NetCache-3.3: 37 mbits/s

Aggregating Cache Nodes Currently have two technologies for clustering cahe nodes L4 Switches Alteon, Arrowpoint, Foundry, and RND WCCP Cisco routers and WCCP licensees