Breaking the on-chip latency barrier using SMART
Authors:
- 1 February 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15300897,p. 378-389
Abstract
As the number of on-chip cores increases, scalable on-chip topologies such as meshes inevitably add multiple hops in each network traversal. The best we can do right now is to design 1-cycle routers, such that the low-load network latency between a source and destination is equal to the number of routers + links (i.e. hops×2) between them. OS/compiler and cache coherence protocols designers often try to limit communication to within a few hops, since on-chip latency is critical for their scalability. In this work, we propose an on-chip network called SMART (Single-cycle Multi-hop Asynchronous Repeated Traversal) that aims to present a single-cycle data-path all the way from the source to the destination. We do not add any additional fast physical express links in the data-path; instead we drive the shared crossbars and links asynchronously up to multiple-hops within a single cycle. We design a router + link microarchitecture to achieve such a traversal, and a flow-control technique to arbitrate and setup multi-hop paths within a cycle. A place-and-routed design at 45nm achieves 11 hops within a 1GHz cycle for paths without turns (9 for paths with turns). We observe 5-8X reduction in low-load latencies across synthetic traffic patterns on an 8×8 CMP, compared to a baseline 1-cycle router. Full-system simulations with SPLASH-2 and PAR-SEC benchmarks demonstrate 27/52% and 20/59% reduction in runtime and EDP for Private/Shared L2 designs.Keywords
This publication has 24 references indexed in Scilit:
- DSENT - A Tool Connecting Emerging Photonics with Electronics for Opto-Electronic Networks-on-Chip Modeling2012
- The IBM Blue Gene/Q Compute ChipIEEE Micro, 2011
- CNoC: High-Radix Clos Network-on-ChipIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2011
- Rigel: A 1,024-Core Single-Chip Accelerator ArchitectureIEEE Micro, 2011
- A 48-Core IA-32 Processor in 45 nm CMOS Using On-Die Message-Passing and DVFS for Performance and Power ScalingIEEE Journal of Solid-State Circuits, 2010
- Express Cube Topologies for on-Chip Interconnects2009
- On-Chip Interconnection Architecture of the Tile ProcessorIEEE Micro, 2007
- A 5-GHz Mesh Interconnect for a Teraflops ProcessorIEEE Micro, 2007
- The BlackWidow High-Radix Clos Network2006
- Microarchitecture of a High-Radix Router2005