Download presentation
Presentation is loading. Please wait.
Published byJuliana Atkinson Modified over 9 years ago
1
1 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) CPRE 583 Reconfigurable Computing Lecture 5: Wed 9/8/2010 (Reconfigurable Computing Architectures) Instructor: Dr. Phillip Jones (phjones@iastate.edu) Reconfigurable Computing Laboratory Iowa State University Ames, Iowa, USA http://class.ece.iastate.edu/cpre583/
2
2 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Chapter 2 (Reconfigurable Architectures) Overview
3
3 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Common Questions
4
4 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Common Questions
5
5 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Common Questions
6
6 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Basic trade-offs associated with different aspects of a Reconfigurable Architecture. (Chapter 2) What you should learn
7
7 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Reconfigurable Architectures Main Idea Chapter 2’s author wants to convey –Applications often have one or more small computationally intense regions of code (kernels) –Can these kernels be sped up using dedicated hardware? –Different kernels have different needs. How does a kernels requirements guide design decisions when implementing a Reconfigurable Architecture?
8
8 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Reconfigurable Architectures Forces that drive a Reconfigurable Architecture –Price Mass production 100K to millions Experimental 1 to 10’s –Granularity of reconfiguration Fine grain Course Grain –Degree of system integration/coupling Tightly Loosely All are a function of the application that will run on the Architecture
9
9 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Example Points in (Price,Granularity,Coupling) Space Price $100’s $1M’s Granularity Coarse Fine Coupling Loose Tight Intel / AMD Int float RFU Processor PC ML507 Ethernet Decode Exec Store
10
10 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) What’s the point of a Reconfigurable Architecture Performance metrics –Computational Throughput Latency –Power Total power dissipation Thermal –Reliability Recovery from faults Increase application performance!
11
11 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Typical Approach for Increasing Performance Application/algorithm implemented in software –Often easier to write an application in software Profile application (e.g. gprof) –Determine where the application is spending its time Identify kernels of interest –e.g. application spends 90% of its time in function matrix_multiply() Design custom hardware/instruction to accelerate kernel(s) –Analysis to kernel to determine how to extract fine/coarse grain parallelism (does any parallelism even exist?) Amdahl’s Law!
12
12 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Amdahl’s Law: Example Application My_app –Running time: 100 seconds –Spends 90 seconds in matrix_mul() What is the maximum possible speed up of My_app if I place matrix_mul() in hardware? What if the original My_app spends 99 seconds in matrx_mul()? 10 seconds = 10x faster 1 seconds = 100x faster Good recent FPGA paper that illustrates increasing an algorithm’s performance with Hardware “NOVEL FPGA BASED HAAR CLASSIFIER FACE DETECTION ALGORITHM ACCELERATION”, FPL 2008 http://class.ece.iastate.edu/cpre583/papers/Shih-Lien_Lu_FPL2008.pdf
13
13 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity
14
14 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Coarse Grain rDPA: reconfigurable Data Path Array Function Units with programmable interconnects ALU Example
15
15 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Coarse Grain rDPA: reconfigurable Data Path Array Function Units with programmable interconnects ALU Example
16
16 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Coarse Grain rDPA: reconfigurable Data Path Array Function Units with programmable interconnects ALU Example
17
17 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Fine Grain FPGA: Field Programmable Gate Array Sea of general purpose logic gates CLB Configurable Logic Block
18
18 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Fine Grain FPGA: Field Programmable Gate Array Sea of general purpose logic gates CLB Configurable Logic Block
19
19 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Fine Grain FPGA: Field Programmable Gate Array Sea of general purpose logic gates CLB Configurable Logic Block
20
20 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Microprocessor
21
21 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Microprocessor 4 3 3 A B op 3
22
22 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Microprocessor 4 3 3 A B op 3 4 3 3 A B 3 4 3 3 A B 3
23
23 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Microprocessor 4 3 3 A B op 3 4 3 3 A B 3 3 3 3 A B
24
24 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Microprocessor 4 3 3 A B op 3
25
25 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Bit logic and constants
26
26 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Bit logic and constants (A and “1100”) or (B or “1000”)
27
27 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Bit logic and constants (A and “1100”) or (B or “1000”) A B
28
28 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Trade-offs Trade-offs associated with LUT size Example: 2-LUT (4=2x2 bits) vs. 10-LUT (1024=32x32 bits) 1024-bits 2-LUT 10-LUT Bit logic and constants (A and “1100”) or (B or “1000”) A AND OR 1 0 B 4 4 It’s much worse, each 10-LUT only has one output Area that was required using 2-LUTS
29
29 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: Example Architectures Fine grain: GARP Course grain: PipeRench
30
30 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: GARP CPU RFU Garp chip Memory I-cache D-cache Config cache
31
31 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: GARP CPU RFU Garp chip Memory I-cache D-cache Config cache RFU control (1) Execution (16, 2-bit) N PE (Processing Element)
32
32 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: GARP CPU RFU Garp chip Memory I-cache D-cache Config cache RFU control (1) Execution (16, 2-bit) N PE (Processing Element) Example computations in one cycle A<<10 | (b&c) (A-2*b+c)
33
33 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: GARP CPU RFU Garp chip Memory I-cache D-cache Config cache Impact of configuration size 1 GHz bus frequency 128-bit memory bus 512Kbits of configuration size On a RFU context switch how long to load a new full configuration? 4 microseconds An estimate of amount of time for the CPU perform a context switch is ~5 microseconds ~2x increase context switch latency!!
34
34 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: GARP CPU RFU Garp chip Memory I-cache D-cache Config cache RFU control (1) Execution (16, 2-bit) N PE (Processing Element) “The Garp Architecture and C Compiler” http://www.cs.cmu.edu/~tcal/IEEE-Computer-Garp.pdf
35
35 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench Coarse granularity Higher (higher) level programming Reference papers PipeRench: A Coprocessor for Streaming Multimedia Acceleration (ISCA 1999): http://www.cs.cmu.edu/~mihaib/research/isca99.pdfhttp://www.cs.cmu.edu/~mihaib/research/isca99.pdf PipeRench Implementation of the Instruction Path Coprocessor (Micro 2000): http://class.ee.iastate.edu/cpre583/papers/piperench_Micro_2000. pdf http://class.ee.iastate.edu/cpre583/papers/piperench_Micro_2000. pdf
36
36 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench Interconnect 8-bit ALU Reg file PE 8-bit ALU Reg file PE 8-bit ALU Reg file PE Interconnect 8-bit ALU Reg file PE 8-bit ALU Reg file PE 8-bit ALU Reg file PE 8-bit ALU Reg file PE 8-bit ALU Reg file PE 8-bit ALU Reg file PE Global bus
37
37 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE Cycle Pipeline stage 1 2 3 4 5 6 0123401234
38
38 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234
39
39 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1
40
40 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2
41
41 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3
42
42 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4
43
43 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4
44
44 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4 Cycle Pipeline stage 1 2 3 4 5 6 012012
45
45 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4 0 Cycle Pipeline stage 1 2 3 4 5 6 012012
46
46 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4 0 Cycle Pipeline stage 1 2 3 4 5 6 012012 0 1
47
47 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4 0 Cycle Pipeline stage 1 2 3 4 5 6 012012 0 1 0 1 2
48
48 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4 0 Cycle Pipeline stage 1 2 3 4 5 6 012012 0 1 0 1 2 3 1 2
49
49 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4 0 Cycle Pipeline stage 1 2 3 4 5 6 012012 0 1 0 1 2 3 1 2 3 4 2
50
50 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench PE 0 Cycle Pipeline stage 1 2 3 4 5 6 0123401234 0 1 0 1 2 1 2 3 2 3 4 0 3 4 0 Cycle Pipeline stage 1 2 3 4 5 6 012012 0 1 0 1 2 3 1 2 3 4 2 3 4 0
51
51 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Independent Reconfigurable Coprocessor –Reconfigurable Fabric does not have direct communication with the CPU Processor + Reconfigurable Processing Fabric –Loosely coupled on the same chip –Tightly coupled on the same chip
52
52 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Main Memory CPU Fetch Decode Execute Memory Write Back L1 Cache L2 Cache Memory Controller DMA Controller I/O Controller USB PCI PCI-ExpressSATA Hard Drive NIC ALU FPU
53
53 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Main Memory CPU Fetch Decode Execute Memory Write Back L1 Cache L2 Cache Memory Controller DMA Controller I/O Controller USB PCI PCI-ExpressSATA Hard Drive NIC ALU FPU RPF
54
54 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Main Memory CPU Fetch Decode Execute Memory Write Back L1 Cache L2 Cache Memory Controller DMA Controller I/O Controller USB PCI PCI-ExpressSATA Hard Drive NIC ALU FPU RPF
55
55 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Main Memory CPU Fetch Decode Execute Memory Write Back L1 Cache L2 Cache Memory Controller DMA Controller I/O Controller USB PCI PCI-ExpressSATA Hard Drive NIC ALU FPU RPF Config I/F
56
56 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Main Memory CPU Fetch Decode Execute Memory Write Back L1 Cache L2 Cache Memory Controller DMA Controller I/O Controller USB PCI PCI-ExpressSATA Hard Drive NIC ALU FPU RPF Config I/F
57
57 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Main Memory CPU Fetch Decode Execute Memory Write Back L1 Cache L2 Cache Memory Controller DMA Controller I/O Controller USB PCI PCI-ExpressSATA Hard Drive NIC ALU FPU RPF I/O Config I/F
58
58 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Degree of Integration/Coupling Main Memory CPU Fetch Decode Execute Memory Write Back L1 Cache L2 Cache Memory Controller DMA Controller I/O Controller USB PCI PCI-ExpressSATA Hard Drive NIC ALU FPU RFU
59
59 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Friday: State machines
60
60 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Questions/Comments/Concerns
61
61 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Lecture 3 notes / slides in progress
62
62 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: PipeRench Scheduling virtual stage on to physical Partial/Dynamically reconfig (each cycle)
63
63 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Granularity: GARP Impact of configuration size on performance Context switching Garp feature Dynamic reconfigurable Store multiple configurations in an on chip cache (4) One configuration at a time Example app mapping to GARP (loop) Amdahl's Law The Garp Architecture and C Compiler http://www.cs.cmu.edu/~tcal/IEEE-Computer-Garp.pdf
64
64 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Overview Dimensions –Price –Granularity –Coupling –To optimize App Performance (compute (throughput, latency), Power, reliability) RPF to efficiently implement VICs –Main picture authors' wants to convey What’s the point or having a Reconfigure arch –Example (Increase App performance) App -> SW/CPU Profile ID kernels of intense compute Design custom hardware/instruction (Amdels law) –Intel FPL paper, great example for reading by Friday
65
65 - CPRE 583 (Reconfigurable Computing): Reconfigurable Computing Architectures Iowa State University (Ames) Reconfigurable Architectures RPF -> VIC (short slide)
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.