Abstract
Bubble flow control is a flow control mechanism commonly used in torus network. Bubble flow control ensures the orderly flow of packets in the ring by efficiently allocating intra-ring buffer resources and solves the intra-ring deadlock problem. However, inter-ring communication with bubble flow control is subject to strict limitations. When packets are communicated across the rings, starvation may occur, resulting in prolonged blocking within the ring. Inefficient inter-ring communication not only generates Head-of-Line blocking, but also affects the whole intra-ring communication through the backpressure mechanism. We propose a Transit Ring design that establishes an alternate ring link in the torus network by reusing free buffers in the router. By evacuating packets blocking inter-ring communication into the Transit Ring, starvation, as well as Head-of-Line blocking, in the network is eliminated. Compared to state-of-the-art bubble flow control, our proposed design improves saturation throughput by an average of 15.9% in a 4\(\times\)4 torus network and 17.06% in an 8\(\times\)8 torus network.
Similar content being viewed by others
Data availability statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
References
Wang B, Lu Z (2020) Advance virtual channel reservation. IEEE Trans Comput 69(9):1320–1334
Wu Y, Wang L, Wang X, Han J, Yin S, Wei S, Liu L (2021) A deflection-based deadlock recovery framework to achieve high throughput for faulty nocs. IEEE Trans Comput Aided Des. Integr. Circuits Syst. 40(10):2170–2183
Rezaei-Ravari M, Naeini VS (2018) Reliable congestion-aware path prediction mechanism in 2D nocs based on efunn. J Supercomput 74(11):6102–6125
Alverson R, Roweth D, Kaplan L (2010) The gemini system interconnect. In: Petrini F, Abts D, Brightwell R, Balaji P, Minkenberg C (eds) IEEE 18th Annual Symposium on High Performance Interconnects, HOTI 2010, Google Campus, Mountain View, California, USA, 18–20Aug 2010, pp 83–87. IEEE Computer Society
Sun C, Ouyang Y, Liang H (2022) Architecting a congestion pre-avoidance and load-balanced wireless network-on-chip. J Parall Distrib Comput 161:143–154
Ma S, Wang Z, Liu Z, Jerger NDE (2015) Leaving one slot empty: Flit bubble flow control for torus cache-coherent nocs. IEEE Trans Comput 64(3):763–777
Chen D, Eisley N, Heidelberger P, Senger RM, Sugawara Y, Kumar S, Salapura V, Satterfield DL, Steinmacher-Burow BD, Parker JJ (2012) The IBM blue gene/q interconnection fabric. IEEE Micro 32(1):32–43
Dally WJ, Towles B (2002) Principles and practices of interconnection networks
Chen L, Wang R, Pinkston TM (2011) Critical bubble scheme: an efficient implementation of globally aware network flow control. In: 25th IEEE International Symposium on Parallel and Distributed Processing, IPDPS 2011, Anchorage, Alaska, USA, 16–20—Conference Proceedings, pp 592–603. IEEE
Tang M, Lin X (2010) Quarter load threshold (qlt) flow control for wormhole switching in mesh-based network-on-chip. J Syst Arch 56(9):452–462
Satish E, Ramachandra A (2022) Comparative performance analysis of routing topology for noc architecture. In: Emerging Research in Computing, Information, Communication and Applications, pp D431–440. Springer
Shikama Y, Kawano R, Matsutani H, Amano H, Nagasaka Y, Fukumoto N, Koibuchi M (2022) A traffic-aware memory-cube network using bypassing. Microprocess Microsyst 90:104471
Ouyang Y, Li R, Hu C, Wang Q, Lu Y, Liang H (2022) Design of fully adaptive routing and hybrid vc allocation in wireless noc. Microelectron J 120:105338
Kwauk G, Kang S, Kasan H, Son H, Kim J (2021) Boomgate: Deadlock avoidance in non-minimal routing for high-radix networks. In:2021 IEEE international symposium on high-performance computer architecture (HPCA), pp 696–708 . IEEE
Dai Y, Lu K, Ma S, Chang J (2022) Full-credit flow control: a novel technique to implement deadlock-free adaptive routing. In:2022 design, automation and test in Europe conference and exhibition (DATE), pp 1041–1046. IEEE
Das S, Karfa C (2022) Deadlock avoidance in torus noc applying controlled move via wraparound channels. In: Artificial Intelligence Driven Circuits and Systems, pp 87–99. Springer
Puente V, Izu C, Beivide R, Gregorio JA, Vallejo F, Prellezo JM (2001) The adaptive bubble router. J Parall Distrib Comput 61(9):1180–1208
Carrión C, Izu C, Gregorio JA, Vallejo F, Beivide R (1998) Ghost packets: a deadlock-free solution for k-ary n-cube networks. In: Proceedings of the Sixth Euromicro Workshop on Parallel and Distributed Processing—PDP ’98, University of Madrid, Madrid, Spain, 21–23 Jan, pp 133–139. IEEE Computer Society
Kermani P, Kleinrock L (1979) Virtual cut-through: a new computer communication switching technique. Comput Netw 3:267–286
Chen L, Pinkston TM (2013) Worm-bubble flow control. In: 19th IEEE International Symposium on High Performance Computer Architecture, HPCA , Shenzhen, China, 23–27 Feb 2013, pp 366–377. IEEE Computer Society
Duato J (1995) A theory of deadlock-free adaptive multicast routing in wormhole networks. IEEE Trans Parall Distrib Syst 6(9):976–987
Puente V, Beivide R, Gregorio JA, Prellezo JM, Duato J, Izu C (1999) Adaptive bubble router: A design to improve performance in torus networks. In: Proceedings of the International Conference on Parallel Processing 1999, ICPP , Wakamatsu, Japan, 21–24 Sept 1999, pp 58–67. IEEE Computer Society
Wang R, Chen L, Pinkston TM (2013) Bubble coloring: avoiding routing- and protocol-induced deadlocks with minimal virtual channel requirement. In: Malony AD, Nemirovsky M, Midkiff SP (eds) International Conference on Supercomputing, ICS’13, Eugene, OR, USA. 10–14 June 2013, pp. 193–202. ACM
Ramrakhyani A, Krishna T (2017) Static bubble: a framework for deadlock-free irregular on-chip topologies. In: 2017 IEEE International Symposium on High Performance Computer Architecture, HPCA 2017, Austin, TX, USA, 4–8 Feb 2017, pp 253–264. IEEE Computer Society
Parasar M, Krishna T (2019) BINDU: deadlock-freedom with one bubble in the network. In: Bogdan P, Silvano C (eds) Proceedings of the 13th IEEE/Acm International Symposium on Networks-on-chip, NOCS 2019, New York, NY, USA, 17–18 Oct 2019, pp 3–138. ACM
Parasar M, Sinha A, Krishna T (2018) Brownian bubble router: enabling deadlock freedom via guaranteed forward progress. In: Twelfth IEEE/ACM International Symposium on Networks-on-Chip, NOCS 2018, Torino, Italy, 4–5 Oct 2018, pp 9–198. IEEE
Fu B, Kim J (2017) Footprint: Regulating routing adaptiveness in networks-on-chip. In: Proceedings of the 44th Annual International Symposium on Computer Architecture, ISCA 2017, Toronto, ON, Canada, 24–28 June 2017, pp 691–702. ACM
Li C, Dong D, Lu Z, Liao X (2018) Rob-router: a reorder buffer enabled low latency network-on-chip router. IEEE Trans Parall Distrib Syst 29(9):2090–2104
Ouyang Y, Yang J, Xing K, Huang Z, Liang H (2018) An improved communication scheme for non-hol-blocking wireless noc. Integral 60:240–247
Parasar M, Jerger NDE, Gratz PV, Miguel JS, Krishna T (2021) SEEC: stochastic escape express channel. In: de Supinski, B.R., Hall, M.W., Gamblin, T. (eds.) SC ’21: The International Conference for High Performance Computing, Networking, Storage and Analysis, St. Louis, Missouri, USA, 14–19 Nov 2021, pp 34–13414. ACM
Ahmed AB, Abdallah AB (2012) LA-XYZ: low latency, high throughput look-ahead routing algorithm for 3d network-on-chip (3d-noc) architecture. In: IEEE 6th International Symposium on Embedded Multicore/manycore SoCs, MCSoC 2012, Fukushima, Japan, 20–22 Sept 2012, pp 167–174. IEEE Computer Society
Peh L, Dally WJ (2001) A delay model and speculative architecture for pipelined routers. In: Proceedings of the Seventh International Symposium on High-Performance Computer Architecture (HPCA’01), Nuevo Leone, Mexico, 20–24 Jan 2001, pp 255–266. IEEE Computer Society
Farrokhbakht H, Kao H, Hasan K, Gratz PV, Krishna T, Miguel JS, Jerger NDE (2021) Pitstop: Enabling a virtual network free network-on-chip. In: IEEE International Symposium on High-Performance Computer Architecture, HPCA 2021, Seoul, South Korea, Feb 27–Mar 3, pp 682–695. IEEE
Song YH, Pinkston TM (2003) A progressive approach to handling message-dependent deadlock in parallel computer systems. IEEE Trans Parallel Distrib Syst 14(3):259–275
Hansson A, Goossens K, Radulescu A (2007) Avoiding message-dependent deadlock in network-based systems on chip. VLSI Design, pp 95859–19585910
Ma S, Jerger NDE, Wang Z (2012) Whole packet forwarding: Efficient design of fully adaptive routing algorithms for networks-on-chip. In: 18th IEEE International Symposium on High Performance Computer Architecture, HPCA 2012, New Orleans, LA, USA, 25–29 pp. 467–478. IEEE Computer Society
Zoni D, Flich J, Fornaciari W (2016) CUTBUF: buffer management and router design for traffic mixing in vnet-based nocs. IEEE Trans Parall Distrib Syst 27(6):1603–1616
Gangwar A, Sreedharan R, Prasad A, Agarwal NK, Gade SH (2021) Topology agnostic virtual channel assignment and protocol level deadlock avoidance in a network-on-chip. In: 58th ACM/IEEE Design Automation Conference, DAC 2021, San Francisco, CA, USA, 5–9 Dec 2021, pp 61–66. IEEE
Mukherjee SS, Bannon PJ, Lang S, Spink A, Webb D (2002) The alpha 21364 network architecture. IEEE Micro 22(1):26–35
Wentzlaff D, Griffin P, Hoffmann H, Bao L, Edwards B, Ramey C, Mattina M, Miao C, III JFB, Agarwal A (2007) On-chip interconnection architecture of the tile processor. IEEE Micro 27(5):15–31
Catania V, Mineo A, Monteleone S, Palesi M, Patti D (2015) Noxim: An open, extensible and cycle-accurate network on chip simulator. In: 26th IEEE International Conference on Application-Specific Systems, Architectures and Processors, ASAP 2015, Toronto, ON, Canada, 27–29 July 2015, pp 162–163. IEEE Computer Society
Bienia C, Kumar S, Singh JP, Li K (2008) The PARSEC benchmark suite: characterization and architectural implications. In: 17th International Conference on Parallel Architectures and Compilation Techniques, PACT 2008, Toronto, Ontario, Canada, 25–29 Oct 2008, pp 72–81. ACM
Acknowledgements
This study was funded by the National Natural Science Foundation of China (NSFC) research Projects (Grant Number 61874157).
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
All authors declare that they have no conflict of interest that are relevant to the content of this article.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Ouyang, Y., Sun, C., Li, R. et al. Transit ring: bubble flow control for eliminating inter-ring communication congestion. J Supercomput 79, 1161–1181 (2023). https://doi.org/10.1007/s11227-022-04712-z
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11227-022-04712-z