1. Communication-Centric Design Robert Mullins Computer Architecture Group Computer Laboratory, University of Cambridge (University of Twente, December 11 th 2006)

2. 2 Convergence to flexible parallel architectures Power Efficient –Better match application characteristics (streaming, coarse-grain parallelism…) –Constraint-driven execution Simple –Increased regularity –S/W programmable –Limited core/tile set –Ease verification issues Flexible –Multi-use platform FPGAs Embedded Processors GPUs Multi-Core Processors SoC Platforms ?

3. 3 Performance from Parallelism Shift to multi-core processors has begun. Performance is boosted in complexity and power effective manner Latencies on-chip are much lower than in previous parallel machines. This should make programming easier + new oppurtinities Easy to increase number of cores as underlying fabrication technology scales Simply an engineering problem?

4. 4 Multi-core Showstoppers! As the number of cores increase current interconnection network designs would require an unacceptable fraction of the chip power budget Attempts at exploiting low-latency on-chip communication are mitigated by complex network interfaces (old problem) and routers Today’s programming models provide little hope of efficiently exploiting a large degree of on-chip parallelism (for the average programmer) – related to N.I. problem Power constraints will be such that only a small percentage of the transistors may be active at once! Why are 100’s of identical cores useful? New interconnection network solutions are critical to making multi- and many-core chips work Power limits imply heterogeneity in some form?

5. 5 Our Group’s Research Now: support evolution of existing platforms –Low-latency and low-power on-chip networks –System-timing considerations –Networking communications within FPGAs –Flexible networked SoC systems, virtual IP –On-chip serial interconnects –Multi-wavelength optical communication (off-chip) –Fault tolerant design Future: –Networks of processors to processing networks –Processing Fabrics FPGAs Embedded Processors GPUs Multi-Core Processors SoC Platforms ?

6. 6 Low-Latency Virtual-Channel Packet-Switched Routers Goal was to develop a virtual-channel network for a tiled processor architecture Collaboration with Krste Asanović’s SCALE group at MIT Problem faced is rising interconnect costs Networking communications can increase communication latencies by an order of magnitude or more!

7. 7 The Lochside test chip (2004/5) UMC 0.18um Process 4x4 mesh network, 25mm 2 Single Cycle Routers (router + link = 1 clock) May be clocked by both traditional H-tree and DCG 4 virtual-channels/input 80-bit links –64-bit data + 16-bit control 250MHz (worst-case PVT) 16Gb/s/channel (~35 FO4) Approx 5M transistors TILE Traffic Generator, Debug & Test R Mullins, West and Moore (ISCA’04, ASP-DAC’06)

8. 8 Virtual-Channel Flow Control

9. 9 Typical Router Pipeline Router pipeline depth limits minimum latency –Even under low traffic conditions –Can make packet buffers less effective –Incurs pipelining overheads

10. 10 Speculative Router Architecture VC and switch allocation may be performed concurrently: –Speculate that waiting packets will be successful in acquiring a VC –Prioritize non-speculative requests over speculative ones Li-Shiuan Peh and William J. Dally, “A Delay Model and Speculative Architecture for Pipelined Routers”, In Proceedings HPCA’01, 2001.

11. 11 Single Cycle Speculative Router

12. 12

13. 13

14. 14 Single Cycle Router Architecture Once speculation mechanism is in place a range of accuracy/cycle-time trade-offs can be made –Blocked VC, pipeline and speculate – use low priority switch scheduler –Switch and VC next request calculation Don’t bother calculating next switch requests just use current set. Safe to be pessimistic about what has been granted. Need to be more accurate for VC allocation –Abort logic accuracy

15. 15 Single Cycle Router Architecture Decreasing accuracy often leads to poorer schedule and more aborts but reduces the router’s cycle time Impact of speculation on single cycle router: –10% more cycles on average –clock period reduced by factor of 1.6 –Network latency reduced by a factor of ~1.5 Need to be careful about updating arbiter state correctly after speculation outcome is known

16. 16 Lochside Router Clock Period 100% standard cell FF/Clocking 23% (8.3 FO4) FIFOs/Control/Datapath53% (19 FO4) Link22% (7.9 FO4) range 4.6-7.9 Could move to router/link pipeline Option to pipeline control - maintaining single cycle best case Impact of technology scaling Scalability: doubling VCs to 8, only adds ~10% to cycle time 30-35 FO4 delays (~800MHz) 5-port router 4 VCs per port 64-bit links, ~1.5mm 90nm technology

17. 17 Router Power Optimisation Local and global clock gating & signal gating –Global clock gating exploits early-request signals from neighbouring routers –Slightly pessimistic (based on what is requested not granted) –Factor 2-4 reduction power consumption Peak 0.15mW/Mhz (0.35 unopt.) Low Random 0.06mW/Mhz (0.27 unopt.) Mullins, SoC’06

18. 18 22% Static power 11% Inter-Router Links ~1% Global Clock tree 65% Dynamic Power –Power Breakdown ~50% local clock tree and input FIFOs ~30% on router datapath ~20% on scheduling and arbitration (Low random traffic case) Analysis of Power Consumption } Due to increase as % as technology scales

19. 19 Distributed Clock Generator (DCG) Exploits self-timed circuitry to generate a clock in a distributed fashion Low-skew and low- power solution to providing global synchrony Mesh topology Simple proof of concept provided by Lochside test chip S. Fairbanks and S. Moore “Self-timed circuitry for global clocking”, ASYNC’05

20. 20 Beyond global synchrony Clock distribution issues –Challenge as network is physically distributed Increasing process variation Synchronization –Core clock frequencies may vary, perhaps adaptively –Link and router DVS or other energy/perf. trade-offs Selecting a global network clock frequency –Run at maximum frequency continuously? –Use a multitude of network clock frequencies? –Select a global compromise?

21. 21 Beyond Global Synchrony Synchronous Delay Insensitive GlobalNone Timing Assumptions Local Relative Wire Delay Less Detection Sub-System Local Isochronic Forks Multiple clocks Data-Driven and Pausible Clocks Bundled Data Quasi-Delay Insensitive Local Clocks, Interaction with data (becoming aperiodic) A complete spectrum of approaches to system-timing exist

22. 22 Data-Driven and Pausible Clocks Mullins/Moore, ASYNC’07

23. 23 Example: AsAP project (UC Davis, 2006) Yu et al, ISSCC’06

24. 24 Example: MAIA chip (Berkeley, 2000) GALS architecture, data-flow driven processing elements (“satellites”) Zhang et al, ISSCC’00

25. 25 Data-Driven Clocking for On-Chip Routers Router should be clocked when one or more inputs are valid (or flits are buffered) Free running (paternoster) elevator –Chain of open compartments –Must synchronise before you jump on! Traditional elevator –Wait for someone to arrive –Close doors, decide who is in and who is out –Metastability issue again (potentially painful!)

26. 26 Data-Driven Clock Implementation Local Clock Generator Template Sample inputs when at least one input is ready (and clock is low) Assert Lock Either admitted or locked out (Close Lift Doors) Incoming data

27. 27 Data-driven clocking benefits Self-timed power gating? DI barrier synchronisation and scheduling extensions NO GLOBAL CLOCK

28. 28 Networks of processors to processing networks How will on-chip networks and cores evolve over time? FPGAs Embedded Processors GPUs Multi-Core Processors SoC Platforms ?

29. 29 Evolving towards processing networks –Network of Processors –Number of processors increase –Core architectures tailored to “many-core” environment –Remove hard tile boundaries Why fix granularity of cores, communication and memory hierarchies? –Move away from processor + router model Everything is on the network, we need much thinner network interfaces Richer interconnection of components, increased flexibility –Add network-based services Network aids collaboration, focuses resources, supports dynamic optimisations, scheduling, … Tailor virtual architecture to application –Processing Network or Fabric Increased flexibility to optimise computation and communication

30. 30 Thank You. 2006 Workshop on On- and Off-Chip Interconnection Networks for Multicore Systems http://www.ece.ucdavis.edu/~ocin06/ (Videos of talks and working group feedback are on-line) http://www.ece.ucdavis.edu/~ocin06/ Computer Architecture: A Quantitative Approach, 4th Edition, Appendix E, entitled Interconnection Networks. http://ceng.usc.edu/smart/slides/appendixE.htmlComputer Architecture: A Quantitative Approach http://ceng.usc.edu/smart/slides/appendixE.html Principles and Practices of Interconnection Networks by William James Dally, Brian Patrick Towles On-chip network bibliography and resources: http://www.cl.cam.ac.uk/~rdm34/onChipNetBib/noc.html http://www.cl.cam.ac.uk/~rdm34/onChipNetBib/noc.html