comparative evaluation of fpga and asic implementations of bufferless and buffered routing...

20
Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie Mellon University

Upload: prosper-richard

Post on 18-Jan-2016

217 views

Category:

Documents


3 download

TRANSCRIPT

Page 1: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered

Routing Algorithms for On-Chip Networks

Yu Cai Ken Mai Onur Mutlu

Carnegie Mellon University

Page 2: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Motivation

In many-core chips, on-chip interconnect (NoC) consumes significant power.

Intel Terascale: ~30%; MIT RAW: ~40% of system power

Must devise low power routing mechanisms

2

Core L1

L2 Slice Router

Page 3: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Destination

Bufferless Deflection Routing Key idea: Packets are never buffered in the network. When two

packets contend for the same link, one is deflected. (BLESS [Moscibroda, ISCA 2009])

3

No buffers lower power, smaller area

Conceptually simple

New traffic can be injectedwhenever there is a freeoutput link.

New traffic can be injectedwhenever there is a freeoutput link.

Page 4: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Motivation

Previous work has evaluated bufferless deflection routing in software simulator (BLESS [Moscibroda, ISCA 2009])

Energy savings Area reduction Minimal performance loss Unfortunately: No real silicon implementation

comparison Silicon power, area, latency, and performance?

Goal: Implement BLESS in FPGA for apple to apple comparison with other buffered routing algorithms.

4

Page 5: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Bufferless Router

Pipeline the oldest first algorithm into two stages Rank priority and oldest first arbiter

5

Page 6: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Oldest-First Arbiter

6

Latency of arbiter cells linearly correlated with port number Latency of each arbiter cells also correlated with port number

Page 7: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Bufferless Router with Buffers

Force schedule can force the flits to be assigned an output port, although it is not the desired port

7

Oldest-firstArbiter

SmartCrossbar

PIPELINE

REGS

PIPELINE

REGS

PIPELINE

REGS

RouteConfiguration

Ages

Highest Priority

Lowest Priority

WN

ES

ReRank Priorities

RouteComputing

Controller(FSM)

Buff er

...

ForceSchedule

ForceSchedule

Feedback

Page 8: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Baseline buffered Router

8

G R O P

VC0

VC1

RouteCompute

VC Req/Rel

VCResponse

SwitchAllocator

VirtualChannelAllocator

Switch Req

Switch Res

Flits

Credits Credits

Flits

WestNorth

EastSouth

Side

Input Controller CrossBar

Router

ResourceLogic

Flits Credits

West

North

South

East

Side

Page 9: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Emulation Platform Setup FPGA platform (Berkeley Emulation Enginee II)

Virtext2Pro FPGA Network topology

3x3 and 4x4 Mesh network 3x3 and 4x4 Torus network

9

Router0000

Router0100

Router1000

Router1100

Router0001

Router0101

Router1001

Router1101

Router0010

Router0110

Router1010

Router1110

Router0011

Router0111

Router1011

Router1111

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

(b) 4x4 Mesh Network

x

y

Router0000

Router0100

Router1000

Router1100

Router0001

Router0101

Router1001

Router1101

Router0010

Router0110

Router1010

Router1110

Router0011

Router0111

Router1011

Router1111

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

Res

(a) 4x4 Torus Network

x

y

Page 10: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Emulation Configuration Router

Buttered router with 1, 2 and 4 virtual channels Bufferless router Bufferless router with one flit buffer in the input port

Traffic patterns

10

Page 11: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

FPGA Area Comparison

11

Page 12: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

FPGA Frequency Analysis

12

Page 13: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

FPGA Power Analysis

13

Page 14: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Network Power Analysis

14

Page 15: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Packet Latency (4x4 Torus)

15

Page 16: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Packet Latency (4x4 Mesh)

16

Page 17: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

ASIC Implementation results

17

Page 18: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Conclusion First realistic and detailed implementation of deflection-

based bufferless routing in on-chip networks

Bufferless routing is efficiently implementable in mesh and torus based on-chip networks, leading to significant area (38%), power consumption (30%) reductions over the best buffered router implementation on a 65nm ASIC design

We identify that oldest-first arbitration, which is used to guarantee livelock freedom of bufferless routing algorithms, as a key design challenge of bufferless router design With careful design, oldest-first arbitration can be

pipelined and can outperform virtual channel arbitration

18

Page 19: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Thank You.

Page 20: Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered Routing Algorithms for On-Chip Networks Yu Cai Ken Mai Onur Mutlu Carnegie

Comparative Evaluation of FPGA and ASIC Implementations of Bufferless and Buffered

Routing Algorithms for On-Chip Networks

Yu Cai Ken Mai Onur Mutlu

Carnegie Mellon University