1. Parallel Computers Organizations and Architecture Department of Computer Science Southern Illinois University Edwardsville Summer, 2015 Dr. Hiroshi Fujinoki E-mail: hfujino@siue.edu CS 312 Computer Organization and Architecture

2. Mult_Sched/001 CS 312 Computer Organization and Architecture Four hardware architecture for “parallel computers”  Tightly-Coupled Multi-Processor System  Functionally-Specialized Multi-Processor System  Loosely-Coupled Multi-Processor System  Distributed Systems (“most loosely coupled systems”)

3. Motherboard Mult_Sched/002  Tightly-Coupled Multi-Processor System Multi-Processor System (multi-processor motherboard) Single-Processor System with a multi-core processor Multi-Processor System Single-Processor System with multi-core processor Processor Processor Core (ALU and others) CS 312 Computer Organization and Architecture

4. Mult_Sched/002  Tightly-Coupled Multi-Processor System Multi-Processor System (multi-processor motherboard) CS 312 Computer Organization and Architecture Two processors on a motherboard

5. Mult_Sched/002  Tightly-Coupled Multi-Processor System CS 312 Computer Organization and Architecture Single-Processor System with a multi-core processor CPU cores

6. Motherboard Graphic Interface Video RAM (“VRAM”) Mult_Sched/003  Functionally-Specialized Multi-Processor System Examples: GPU on graphics card Built-in processor on high-speed disk controllers or NICs (especially those using DMA) Processor Monitor (CRT, Flat Panel) DAC  Graphic-card performs D/A conversion using DAC.  GPU  GPU processes image data in the graphic-card memory   Processor sends graphic command to GPU   Graphic-card sends analog image signals (RGB-signals) to monitor  (GPU = “Graphic Processing Unit”) CS 312 Computer Organization and Architecture

7. Mult_Sched/003  Functionally-Specialized Multi-Processor System Examples: GPU on graphics card (GPU = “Graphic Processing Unit”) CS 312 Computer Organization and Architecture DMA SCSI I/O card CPU Control Program (in ROM)

8. Mult_Sched/004  Loosely-Coupled Multi-Processor System Multi-Systemboard (multiple motherboard) computers Computer System “Bus” Processor System Board (Motherboard) Memory A computer with multiple motherboards (“blades”) Blades communicate through the bus Each blade is a computer Communication delay over the bus at least “  s” order CS 312 Computer Organization and Architecture

9. Mult_Sched/004  Loosely-Coupled Multi-Processor System Multi-Systemboard (multiple motherboard) computers CS 312 Computer Organization and Architecture

10. Mult_Sched/005  Distributed Systems (“most loosely coupled systems”) AS 1 AS 4 AS 2 AS 3 Processor Local Memory Secondary Storage Other I/O Processor Local Memory Secondary Storage Other I/O Processor Local Memory Secondary Storage Other I/O Processor Local Memory Secondary Storage Other I/O Process (executable codes) Process Migration File (data) Data Migration Network CS 312 Computer Organization and Architecture

11. Mult_Sched/006 Three different types of tightly-coupled multi-processor systems (1) “Fine-grained” multi-processor parallel computers (2) “Medium-grained” multi-processor parallel computers (3) “Coarse-grained” multi-processor parallel computers CS 312 Computer Organization and Architecture

12. Mult_Sched/007 Fine-Grained Multi-Process Fine-grained = instruction-level multi-processing Your program (binary executable) A = B + C; X = Y + Z; W = A + X; synchronization Dependency Granularity: 1~20 instructions CPU CS 312 Computer Organization and Architecture

13. Mult_Sched/008 Medium-Grained Multi-Process Medium-grained = thread-level multi-processing Your program (binary executable) Thread A Thread B Thread C Thread D Processor CS 312 Computer Organization and Architecture

14. Mult_Sched/009 Medium-Grained Multi-Process Example: Web Browser Thread A -- Display thread (text output & jpeg image processing) Thread B -- Taking user inputs (edit boxes, radio boxes in the browser window Thread C -- Network input (receiving data from network) Thread D -- Network output (sending data to network) Thread A Thread B Thread C Thread D Receiving data Displaying data User makes inputs Receiving data Transmit data CS 312 Computer Organization and Architecture

15. Mult_Sched/010 Medium-Grained Multi-Process Example: Web Browser Thread A -- Display thread (text output & jpeg image processing) Thread B -- Taking user inputs (edit boxes, radio boxes in the browser window Thread C -- Network input (receiving data from network) Thread D -- Network output (sending data to network) Thread A Thread B Thread C Thread D Receiving data Displaying data User makes inputs Receiving data Transmit data Browser execution with better responses Granularity: 20~200 instructions CS 312 Computer Organization and Architecture

16. Mult_Sched/011 Coarse-Grained Multi-Process Coarse-grained = process-level multi-tasking Process assignment to multiple processors in multi-tasking environment Memory Processor Time CS 312 Computer Organization and Architecture

17. Mult_Sched/012 Coarse-Grained Multi-Process Coarse-grained = process-level multi-tasking Process assignment to multiple processors in multi-tasking environment Memory Processor Pool Granularity = ms order 1ms (@ 1GHz) = 1 million instructions 100ms (@ 1GHz) = 100M instructions Granularity: 1~100 M instructions Time CS 312 Computer Organization and Architecture