Token Coherence: Decoupling Performance and Correctness PowerPoint PPT Presentation

presentation player overlay
About This Presentation
Transcript and Presenter's Notes

Title: Token Coherence: Decoupling Performance and Correctness


1
Token Coherence Decoupling Performance and
Correctness
  • Milo M. D. Martin
  • Mark D. Hill
  • David A. Wood
  • University of Wisconsin-Madison
  • ISCA-30 (2003)

2
Outline
  • Token Coherence Basics
  • What Is Token Coherence?
  • What Are Its Advantages?
  • Review of Snooping and Directory-Based Coherence
    Schemes
  • Token Coherence Details
  • Evaluation Results and Conclusions

3
Token Coherence Basics
  • Decouple Interconnect Performance and Protocol
    Correctness
  • Remember Amdahls Law Go Faster In Most Cases At
    Expense Of Occasional, More Expensive Corner Case
    Handling
  • Fast, Unordered Bus For Protocol Traffic
  • Races Allowed, Handling Built-In, But Deferred
  • Cache Lines Have Tokens To Track State
  • Token Ownership Correlates With Coherence State
    (MOSI)

4
Token Coherence Benefits
  • Fast Bus Makes Common Case Fast
  • Some Additional Bus Traffic (Not Too Much)
  • Common Is Very Common (95)
  • Benefits Of Snooping And Directories
  • Unordered Bus Messages (Directories)
  • Cache-Cache Transfers Without Indirection
    (Snooping)

5
Outline
  • Token Coherence Basics
  • What Is Token Coherence?
  • What Are Its Advantages?
  • Review of Snooping and Directory-Based Coherence
    Schemes
  • Token Coherence Details
  • Evaluation Results and Conclusions

6
Coherence Review Snooping
  • Requests Broadcast, All Agents Snoop
  • One-Hop Communication (Low Latency)
  • Every Agent Processes Every Message
  • Bus Is Point Of Synchronization (Total Order)
  • Hard To Scale To Higher Speeds/Larger Systems

Total Order of Events 1 Before 2 Before 3 To All
Bus Agents
P0
P1
P2
P3
1 2 3
1 2 3
1 3
2 3
1
2
Mem
2 1
3
7
Coherence Review Directories
  • Directory Manages Ordering Of Requests
  • Bus Can Be Faster (Unordered)
  • Adds Level Of Indirection To Cache-Cache
    Transactions (Extra Latency)
  • Messages Addressed To Specific Agents

Bus Order Irrelevant, Ordering At Directory 1
Before 2 Before 3
P0
P1
P2
P3
3
3
1
2
3
3
Dir
2 1
3
8
Outline
  • Token Coherence Basics
  • What Is Token Coherence?
  • What Are Its Advantages?
  • Review of Snooping and Directory-Based Coherence
    Schemes
  • Token Coherence Details
  • Evaluation Results and Conclusions

9
Token Coherence Goals
  • Achieve Efficiency Of Snooping
  • Low-Latency Inter-Cache Transfers
  • Keep Advantage Of Directories
  • Unordered, Fast Interconnects Allowed
  • How?
  • Use Of Tokens To Implement Correctness
    Substrate And Protocol(s)
  • Tokens Associated With Each Cache Line
  • At Least N Tokens Per Line (N Processors)

10
Correctness Substrate
  • Enforce Safety Count Tokens
  • Tokens Preserved (Always T Tokens Per Line)
  • Must Hold All Tokens To Write
  • Must Hold One Token To Read
  • If A Message Has A Token, It Must Have Data
  • Optimization Add Special Owner Token
  • Tokens Preserved With One Owner Token Per Line
  • Must Hold All Tokens To Write
  • Must Hold One Token And Valid Data To Read
  • If A Message Has Owner Token, It Must Have Data

11
A Token Coherence Protocol
  • TokenB Token Coherence Using Broadcast
  • Processors Broadcast Transient Token Requests
  • Snooping Mayhem Ensues Without Total Ordering Of
    Transactions On Bus
  • Races Allowed To Occur Unanswered Messages
    Retried
  • Last Resort (After Several Retries) Is
    Aptly-Named Persistent Request

12
Outline
  • Token Coherence Basics
  • What Is Token Coherence?
  • What Are Its Advantages?
  • Review of Snooping and Directory-Based Coherence
    Schemes
  • Token Coherence Details
  • Evaluation Results and Conclusions

13
Performance Evaluation
  • Compared To Snooping, Directories, And Glueless
    Estimated Protocol
  • Simulation Results Show
  • Token Coherence Works, Performs Well, Even
  • For 16-Processor System, gt95 Of First Transient
    Requests Succeed With TokenB
  • Potentially Significant Additional Traffic
    Generated Over Directory
  • Not Ultimately As Scalable As Directories

14
Conclusions And Discussion
  • Token Coherence Provides Novel Solution To
    Snooping/Directory Hybrid Optimization Problem
  • Is Additional Space For Token Storage An Issue?
  • Why Didnt Hammer/21364 Use This?
  • With Glueless A Trend, Is This Doomed To
    Academia?
  • How Does This Compare To Multicast Snooping?
Write a Comment
User Comments (0)
About PowerShow.com