180 likes | 316 Views
Do We Need Wide Flits in Networks-On-Chip?. Junghee Lee, Chrysostomos Nicopoulos , Sung Joo Park, Madhavan Swaminathan and Jongman Kim. Presented by Junghee Lee. Introduction. Increasing number of cores Communication-centric Packet-based Networks-on-Chip Unit
E N D
Do We Need Wide Flits in Networks-On-Chip? Junghee Lee, ChrysostomosNicopoulos, Sung Joo Park, MadhavanSwaminathan and Jongman Kim Presented by Junghee Lee
Introduction • Increasing number of cores Communication-centric Packet-based Networks-on-Chip • Unit • Packet: a meaningful unit of the upper-layer protocol • Flit: the smallest unit of flow control maintained by NoC • If a packet is larger than a flit, a packet is split into multiple flits • The flit size usually matches with the physical channel width
Motivation 256 Intel Sandy Bridge 64 or 128 Research papers 144 Intel Single-Chip Cloud What is the optimal flit sizein Networks-on-Chipforgeneral purpose computing? 256 or 512 Research papers 160 Tilera
Multifaceted Factors A first attempt in drawing balanced conclusion
Assumed NoC Router Architecture d v p c
Packet and Flit Header Payload
#1) Global Wires • Can we afford wide flits as technology scales? Technology scaling does not allow for a direct widening of the flits because the power portion of the global wires increases as technology scales * International Technology Roadmap for Semiconductors (ITRS) 2009 and 2011
#2) Cost of Router • Is the cost of wide-flit routers justifiable? Cost of buffers Flit size Buffer depth Number of virtual channels Cost of switch (Flit size)2 (Number of ports)2 Switch Cost Flit size 2 cost of router 2.97 Flit size 4 cost of router 10.10 Buffer If the performance improvement does not compensate for the increase in the cost, widening of the flit size is hard to justify Flit size
#3) Latency • The network traffic usually consists of packets of different sizes • ls: The size of shortest packet • ll: The size of longest packet • How much do wide flits contribute to overall performance? Latency Suggested rule of thumb: Flit size = shortest packet size + header overhead Flit size ls+h ll+h
#4) Workload Characteristics The injection rate of real applications is far less than the typical saturation point of NoC Self-throttling effect [34] • Do memory-intensive workloads need wide flits? Up to 64 cores, we can keep the rule of thumb because of the low injection rate
#5) Throughput • Widening the flit is not a cost-effective way because of fragmentation • If widening the physical channel is the only option for increasing the throughput, we suggest using physically separated networks Latency • Do we need wider flits as the number of processing elements increases? One 80-bit network One 160-bit network Two 80-bit networks Flit size
Conclusions • No, unless the power budget for NoC increases • No, the cost increases sharply with the flit size • Until the flit size reaches the shortest packet size • No, because of self-throttling effect • No, because of fragmentation
Final Conclusion • Suggested rule of thumb:Flit size = shortest packet size + header overhead • This paper provides a comprehensive discussion on all key aspects pertaining to the NoC’s flit size • This exploration could serve as a quick reference for the designers/architects of general-purpose multi-core microprocessors who need to decide on an appropriate flit size for their design.
Questions? Contact info Junghee Lee junghee.lee@gatech.edu Electrical and Computer Engineering Georgia Institute of Technology