Y.C. Tay National University of Singapore

Slides:



Advertisements
Similar presentations
Abstract There is significant need to improve existing techniques for clustering multivariate network traffic flow record and quickly infer underlying.
Advertisements

Technology Drivers Traditional HPC application drivers – OS noise, resource monitoring and management, memory footprint – Complexity of resources to be.
Multi-Layer Switching Layers 1, 2, and 3. Cisco Hierarchical Model Access Layer –Workgroup –Access layer aggregation and L3/L4 services Distribution Layer.
Project Proposal Presented by Michael Kazecki. Outline Background –Algorithms Goals Ideas Proposal –Introduction –Motivation –Implementation.
Hit or Miss ? !!!.  Small size.  Simple and fast.  Implementable with hardware.  Does not need too much power.  Does not predict miss if we have.
“This research was, in part, funded by the U.S. Government. The views and conclusions contained in this document are those of the authors and should not.
Web Caching Schemes1 A Survey of Web Caching Schemes for the Internet Jia Wang.
6/14/2015 How to measure Multi- Instruction, Multi-Core Processor Performance using Simulation Deepak Shankar Darryl Koivisto Mirabilis Design Inc.
Continuous Process Improvement
CS 1 – Introduction to Computer Science Introduction to the wonderful world of Dr. T Daniel Tauritz, Ph.D. Associate Professor of Computer Science.
Building a Strong Foundation for a Future Internet Jennifer Rexford ’91 Computer Science Department (and Electrical Engineering and the Center for IT Policy)
Variable Speed Limits Binh Dong. 9/16/20112 Problem Many speed limits are underrated Time and traffic has influences on flow Variances in speed causes.
OSPF To route, a router needs to do the following: Know the destination address Identify the sources it can learn from Discover possible.
Towards Autonomic Hosting of Multi-tier Internet Services Swaminathan Sivasubramanian, Guillaume Pierre and Maarten van Steen Vrije Universiteit, Amsterdam,
Inter-domain Routing Outline Border Gateway Protocol.
Internet GIS. A vast network connecting computers throughout the world Computers on the Internet are physically connected Computers on the Internet use.
Rutgers IT Complex Michael R Mundrane 4 December 2001 Rutgers University Computing Services.
Overview Integration Testing Decomposition Based Integration
1 Process Engineering A Systems Approach to Process Improvement Jeffrey L. Dutton Jacobs Sverdrup Advanced Systems Group Engineering Performance Improvement.
Oracle Challenges Parallelism Limitations Parallelism is the ability for a single query to be run across multiple processors or servers. Large queries.
A Data Cache with Dynamic Mapping P. D'Alberto, A. Nicolau and A. Veidenbaum ICS-UCI Speaker Paolo D’Alberto.
IISWC 2007 Panel Analyzing Petabytes Suchi Raman Netezza Corp.
© SKY Computers, Inc. All Rights Reserved 9/25/02 Slide 1 VSIPL, from API to Product Sharon M. Sacco.
Compilers for Embedded Systems Ram, Vasanth, and VJ Instructor : Dr. Edwin Sha Synthesis and Optimization of High-Performance Systems.
Packet Capture and Analysis: An Introduction to Wireshark 1.
Self-Managing Cost Models Shivnath Babu Stanford University.
Memory Hierarchy Adaptivity An Architectural Perspective Alex Veidenbaum AMRM Project sponsored by DARPA/ITO.
IT 606 Computer Networks (CN). 1.Evolution of Computer Networks & Application Layer. 2.Transport Layer & Network Layer. 3.Routing & Data link Layer. 4.Physical.
© 2005 Cisco Systems, Inc. All rights reserved. BGP v3.2—6-1 Scaling Service Provider Networks Scaling IGP and BGP in Service Provider Networks.
Evolvable Hardware Questions What is it? Why do we want it? Who is it for? How do we get it?
Combining Software and Hardware Monitoring for Improved Power and Performance Tuning Eric Chi, A. Michael Salem, and R. Iris Bahar Brown University Division.
1 Sheer volume and dynamic nature of video stresses network resources PIE: A lightweight latency control to address the buffer problem issue Rong Pan,
Meeting 8: Features for Object Classification Ullman et al.
Enhanced Availability With RAID CC5493/7493. RAID Redundant Array of Independent Disks RAID is implemented to improve: –IO throughput (speed) and –Availability.
Virtual-Topology Adaptation for WDM Mesh Networks Under Dynamic Traffic.
© ExplorNet’s Centers for Quality Teaching and Learning 1 Objective % Understand digital graphics production methods, software, and hardware.
تجارت الکترونیک سیار جلسه پنجم مدرس : دکتررامین کریمی.
Презентацию подготовила Хайруллина Ч.А. Муслюмовская гимназия Подготовка к части С ЕГЭ.
Graphical Presentation of Mobile networks
Ioannis E. Venetis Department of Computer Engineering and Informatics
Associativity in Caches Lecture 25
Team Name: OCD Solutions
ENG3050 Embedded Reconfigurable Computing Systems
Graphics: Production Methods, software, & Hardware

Readykoncepts.com.au: Search Engine Marketing Sydney, SEO Brisbane Internet Marketing 
Lecture 13: Large Cache Design I
Working at a Small-to-Medium Business or ISP – Chapter 6
Business Rule Based Configuration Management and Software System Implementation Using Decision Tables Olegas Vasilecas, Aidas Smaizys VGTU, Vilnius, Lithuania.
QuickBooks Payroll Support 1(800) QuickBooks Payroll Customer Service QuickBooks. To provide each possible payroll assistance and help users.
Introduction SWE 619.
Network Optimizer Optimize Your Business & Cloud Networks
Continuous Process Improvement
IS4680 Security Auditing for Compliance
A Cognitive Approach for Cross-Layer Performance Management
أنماط الإدارة المدرسية وتفويض السلطة الدكتور أشرف الصايغ
Jigar.B.Katariya (08291A0531) E.Mahesh (08291A0542)
The Use of IATI Data in Country Systems
John D. McGregor Module 6 Session 1 More Design
Marketing.
Operation Modes NIGHT Background processing Dialog processing DAY BTC
Learning Adaptive profiling Exception profiling Exception heuristics

Identification of Variation Points Using Dynamic Analysis
Dynamic WAN Selection Optimize Your Business & Cloud Networks
Automatic Tuning of Two-Level Caches to Embedded Applications
Objective % Describe digital video production methods.
Internet Exchange.
Group 9 Jeffrey Cook, Derrick Williams Chris Juaquez, Malick Oussieni
Presentation transcript:

Y.C. Tay National University of Singapore Dynamic System Optimization through Performance Modeling: Universality and Decomposition Y.C. Tay National University of Singapore

Dynamic optimization is hard. Examples: Prob(miss) = f(cache size) Given target Prob(miss), how to dynamically adjust cache size? Universality (2) Internet traffic equilibrium How to prevent performance collapse from congestion? Decomposition

(1) Universality reference pattern cache size management policy issue: Prob(miss) = f(cache size) reference pattern cache size management policy complex add: ● system customization ● hw/sw evolution ● autonomic configuration ● dynamic adjustment hardware variation data layout application mix software variation data instance intractable! system configuration

(1) Universality no change idea: Prob(miss) = f(cache size | parameters) change values ● system customization ● hw/sw evolution ● autonomic configuration ● dynamic adjustment

(1) Universality = (H + √(H 2 – 4))(n*+n0) – n0 idea: Prob(miss) = f(cache size | parameters) #miss = f (M | M*, M0, n*, n0) change values = (H + √(H 2 – 4))(n*+n0) – n0 1 2 where H = 1 + M* - M0 M - M0 ● system customization ● hw/sw evolution ● autonomic configuration ● dynamic adjustment

(1) Universality = (H + √(H 2 – 4))(n*+n0) – n0 #miss = f (M | M*, M0, n*, n0) = (H + √(H 2 – 4))(n*+n0) – n0 1 2 where H = 1 + M* - M0 M - M0 Example: cache=RAM