Presentation is loading. Please wait.

Presentation is loading. Please wait.

Mattan Erez The University of Texas at Austin

Published byFarida Hermanto Modified over 5 years ago

Similar presentations

Presentation on theme: "Mattan Erez The University of Texas at Austin"— Presentation transcript:

1 Mattan Erez The University of Texas at Austin
EE382N (20): Computer Architecture - Parallelism and Locality Fall 2011 Lectures 8 & 9 – Parallelism in Hardware Mattan Erez The University of Texas at Austin EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

2 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer memory hierarchy EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

3 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:add r5, r1, r3 3:add r6, r2, r3 1 F D I R E W C 2 3 EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

4 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:add r5, r1, r3 3:add r6, r2, r3 1 F D I R E W C 2 3 EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

5 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:add r5, r1, r3 3:add r6, r2, r3 1 F D I R E W C 2 3 EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

6 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:add r5, r1, r3 3:add r6, r2, r3 1 F D I R E W C 2 3 What are the parallel resources? EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

7 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:add r5, r1, r4 3:add r6, r5, r3 1 F D I R E W C 2 3 EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

8 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:add r5, r1, r4 3:add r6, r5, r3 1 F D I R E W C 2 3 EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

9 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:add r5, r1, r3 3:add r6, r2, r3 1 F D I R E W C 2 3 Communication and synchronization mechanisms? EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

10 Simplified view of a pipelined processor
dispatch (issue) reg. access execute write-back commit fetch decode sequencer 1:add r4, r1, r2 2:ld r5, r4 3:add r6, r5, r3 4:add r7, r1, r3 1 F D I R E W C 2 L 3 4 EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

11 Simplified view of a OOO pipelined processor
dispatch (issue) issue (dispatch) reg. access execute write-back commit fetch decode rename sequencer scheduler 1:add r4, r1, r2 2:ld r5, r4 3:add r6, r5, r3 4:add r7, r1, r3 1 F D I d R E W C 2 L 3 4 Communication and synchronization mechanisms? EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

12 Pipelining Summary Pipelining is using parallelism to hide latency
Do useful work while waiting for other work to finish Multiple parallel components, not multiple instances of same component Examples: Execution pipeline Memory pipelines Issue multiple requests to memory without waiting for previous requests to complete Software pipelines Overlap different software blocks to hide latency: computation/communication EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

13 Parallel Execution Concurrency Communication Synchronization
what are the multiple resources? Communication and storage Synchronization Implicit? Explicit? what is being shared ? What is being partitioned ? EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

14 Resources in a parallel processor/system
Execution ALUs Cores/processors Control Sequencers Instructions OOO schedulers State Registers Memories Networks EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

15 Communication and synchronization
Clock – explicit compiler order Explicit signals (e.g., dependences) Implicit signals (e.g., flush/stall) More for pipelining than multiple ALUs Communication Bypass networks Registers Memory Explicit (over some network) EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

16 Organizations for ILP (for multiple ALUs)
EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

17 Superscalar (ILP for multiple ALUs)
sequencer scheduler Synchronization Explicit signals (dependences) Communication Bypass, registers, mem Shared Sequencer, OOO, registers, memories, net, ALUs Partitioned Instructions memory hierarchy How many ALUs? EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

18 SMT/TLS (ILP for multiple ALUs)
EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

19 SMT/TLS (ILP for multiple ALUs)
sequencer scheduler sequencer Synchronization Explicit signals (dependences) Communication Bypass, registers, mem Shared OOO, registers, memories, net, ALUs Partitioned Sequencer, Instructions, arch. registers memory hierarchy Why is this ILP? How many threads? EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

20 VLIW (ILP for multiple ALUs)
EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

21 VLIW (ILP for multiple ALUs)
sequencer Synchronization Clock+compiler Communication Registers, mem, bypass Shared Sequencer, OOO, registers, memories, net Partitioned Instructions, ALUs memory hierarchy How many ALUs? EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

22 Explicit Dataflow (ILP for multiple ALUs)
EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

23 Explicit Dataflow (ILP for multiple ALUs)
sequencer scheduler scheduler scheduler scheduler Synchronization Explicit signals Communication Registers+explicit Shared Sequencer, memories, net Partitioned Instructions, OOO, ALUs memory hierarchy EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

24 DLP for multiple ALUs From HW – this is SIMD
EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

25 SIMD (DLP for multiple ALUs)
sequencer Synchronization Clock+compiler Communication Explicit Shared Sequencer, instructions Partitioned Registers, memories, ALUs Sometimes: memories, net Local Mem Local Mem Local Mem Local Mem memory hierarchy EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

26 Vectors (DLP for multiple ALUs)
sequencer Local Mem Local Mem Local Mem Local Mem memory hierarchy Vectors: memory addresses are part of single-instruction and not part of multiple-data EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

27 TLP for multiple ALUs EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

28 MIMD – shared memory (TLP for multiple ALUs)
sequencer sequencer sequencer sequencer scheduler scheduler scheduler scheduler Synchronization Explicit, memory Communication Memory Shared Memories, net Partitioned Sequencer, instructions, OOO, ALUs, registers, some nets memory hierarchy EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

29 MIMD – distributed memory
sequencer sequencer sequencer sequencer scheduler scheduler scheduler scheduler Synchronization Explicit Communication Shared Net Partitioned Sequencer, instructions, OOO, ALUs, registers, some nets, memories memory hierarchy memory hierarchy memory hierarchy memory hierarchy EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

30 Summary of communication and synchronization
Style Synchronization Communication Superscalar explicit signals (RS) registers+bypass VLIW clock+compiler registers (bypass?) Dataflow explicit signals registers+explicit SIMD explicit MIMD memory+explicit EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

31 Summary of sharing in ILP HW
Style Seq Inst OOO Regs Mem ALUs Net Superscalar S P SMT/TLS VLIW N/A Dataflow B EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

32 Summary of sharing in DLP and TLP
Style Seq Inst OOO Regs Mem ALUs Net Vector S N/A P s B SIMD p MIMD S/P EE382N: Parallelilsm and Locality, Fall Lectures 8 & 9 (c) Mattan Erez

Download ppt "Mattan Erez The University of Texas at Austin"

Similar presentations

Machine cycle.

Machine cycle.
Lecture 4: CPU Performance

Lecture 4: CPU Performance
Computer Organization and Architecture

Computer Organization and Architecture
Real-time Signal Processing on Embedded Systems Advanced Cutting-edge Research Seminar I&III.

Real-time Signal Processing on Embedded Systems Advanced Cutting-edge Research Seminar I&III.
Multiprocessors— Large vs. Small Scale Multiprocessors— Large vs. Small Scale.

Multiprocessors— Large vs. Small Scale Multiprocessors— Large vs. Small Scale.
Computer Structure 2014 – Out-Of-Order Execution 1 Computer Structure Out-Of-Order Execution Lihu Rappoport and Adi Yoaz.

Computer Structure 2014 – Out-Of-Order Execution 1 Computer Structure Out-Of-Order Execution Lihu Rappoport and Adi Yoaz.
Superscalar processors Review. Dependence graph S1S2 Nodes: instructions Edges: ordered relations among the instructions Any ordering-based transformation.

Superscalar processors Review. Dependence graph S1S2 Nodes: instructions Edges: ordered relations among the instructions Any ordering-based transformation.
Instruction Level Parallelism María Jesús Garzarán University of Illinois at Urbana-Champaign.

Instruction Level Parallelism María Jesús Garzarán University of Illinois at Urbana-Champaign.
CSE 490/590, Spring 2011 CSE 490/590 Computer Architecture VLIW Steve Ko Computer Sciences and Engineering University at Buffalo.

CSE 490/590, Spring 2011 CSE 490/590 Computer Architecture VLIW Steve Ko Computer Sciences and Engineering University at Buffalo.
Dynamic Branch PredictionCS510 Computer ArchitecturesLecture 10 - 1 Lecture 10 Dynamic Branch Prediction, Superscalar, VLIW, and Software Pipelining.

Dynamic Branch PredictionCS510 Computer ArchitecturesLecture Lecture 10 Dynamic Branch Prediction, Superscalar, VLIW, and Software Pipelining.
Instruction-Level Parallelism (ILP)

Instruction-Level Parallelism (ILP)
C SINGH, JUNE 7-8, 2010IWW 2010, ISATANBUL, TURKEY Advanced Computers Architecture, UNIT 1 Advanced Computers Architecture Lecture 4 By Rohit Khokher Department.

C SINGH, JUNE 7-8, 2010IWW 2010, ISATANBUL, TURKEY Advanced Computers Architecture, UNIT 1 Advanced Computers Architecture Lecture 4 By Rohit Khokher Department.
© Kavita Bala, Computer Science, Cornell University Kevin Walsh CS 3410, Spring 2010 Computer Science Cornell University Pipelining See: P&H Chapter 4.5.

© Kavita Bala, Computer Science, Cornell University Kevin Walsh CS 3410, Spring 2010 Computer Science Cornell University Pipelining See: P&H Chapter 4.5.
Computer Architecture 2011 – Out-Of-Order Execution 1 Computer Architecture Out-Of-Order Execution Lihu Rappoport and Adi Yoaz.

Computer Architecture 2011 – Out-Of-Order Execution 1 Computer Architecture Out-Of-Order Execution Lihu Rappoport and Adi Yoaz.
Instructor: Senior Lecturer SOE Dan Garcia CS 61C: Great Ideas in Computer Architecture Pipelining Hazards 1.

Instructor: Senior Lecturer SOE Dan Garcia CS 61C: Great Ideas in Computer Architecture Pipelining Hazards 1.
Instruction Level Parallelism (ILP) Colin Stevens.

Instruction Level Parallelism (ILP) Colin Stevens.
Computer Architecture 2010 – Out-Of-Order Execution 1 Computer Architecture Out-Of-Order Execution Lihu Rappoport and Adi Yoaz.

Computer Architecture 2010 – Out-Of-Order Execution 1 Computer Architecture Out-Of-Order Execution Lihu Rappoport and Adi Yoaz.
CS 300 – Lecture 24 Intro to Computer Architecture / Assembly Language The LAST Lecture!

CS 300 – Lecture 24 Intro to Computer Architecture / Assembly Language The LAST Lecture!
Basics and Architectures

Basics and Architectures
Very Long Instruction Word (VLIW) Architecture. VLIW Machine It consists of many functional units connected to a large central register file Each functional.

Very Long Instruction Word (VLIW) Architecture. VLIW Machine It consists of many functional units connected to a large central register file Each functional.

Similar presentations

Ads by Google