AB∆Zip: An Approximate Base Delta End-to-End Packet Compression Framework for Network-on-Chips

AB∆Zip: An Approximate Base Delta End-to-End Packet Compression Framework for Network-on-Chips
  IJETT-book-cover           
  
© 2022 by IJETT Journal
Volume-70 Issue-6
Year of Publication : 2022
Authors : T. Pullaiah, K. Manjunatha Chari, B.L. Malleswari
DOI : 10.14445/22315381/IJETT-V70I6P222

How to Cite?

T. Pullaiah, K. Manjunatha Chari, B.L. Malleswari, "AB∆Zip: An Approximate Base Delta End-to-End Packet Compression Framework for Network-on-Chips," International Journal of Engineering Trends and Technology, vol. 70, no. 6, pp. 195-208, 2022. Crossref, https://doi.org/10.14445/22315381/IJETT-V70I6P222

Abstract
Approximate communication methods can be applied in different domains, including pattern recognition and data mining, to enhance transmission efficiency in power and delay while guaranteeing tolerable output quality. This paper proposes a new Approximate Base delta (AB∆Zip) packet compression framework. This approach initially truncates the integer or floating-point data based on the error threshold for the reduction of power consumption and network latency. This model uses an approximate level configuration framework to compute the optimal error threshold based on error tolerance. After forming the approximate pattern, a B∆ compression method is used to compress the approximate pattern. Here, the B∆ compression method is modified by identifying the data patterns within a flit and is used to minimize the size of the subtractor components. Also, the proposed method uses frequent pattern compression (FPC) scheme to replace the frequent pattern with the codeword for uncompressible chunks of the delta compression method. It avoids the transmission of a small valued floating-point and integer with a larger number of Most significant bits (MSBs). The simulation results illustrate that the proposed AB∆Zip increases the compression ratio to 3.7% at a 10% error threshold and minimizes the network latency, area, and power consumption to 42.1%, 4.8%, and 11.76%, respectively to the most recent existing approximate communication method.

Keywords
Approximate communication, network on chip (NoC), Base-delta (B∆) compression, error tolerance, and latency.

Reference
[1] Jedidi, Detection and Monitoring Intra/Inter Crosstalk in Optical Network on Chip, International Journal of Electrical & Computer Engineering. 8(6) (2018) 2088-8708.
[2] S. Kashi and A. Patooghy, Row/Column-First: A Path-Based Multicast Algorithm for 2D Mesh-based Network on Chips, Iranian Journal of Electrical and Electronic Engineering. 14(2) (2018) 124-136.
[3] S.S. Kendaganna, A. Jatti and B.V. Uma, Design and Implementation of Secured Agent Based Noc Using Shortest Path Routing Algorithm, International Journal of Electrical and Computer Engineering. 9(2) (2019) 950.
[4] M.F. Reza & P. Ampadu, Approximate Communication Strategies for Energy-Efficient and High Performance Noc: Opportunities and Challenges, In Proceedings of the on Great Lakes Symposium on VLSI. (2019) 399-404.
[5] K. Zhou, Y. He, R. Xiao, J. Liu & K. Huang, A Customized NoC Architecture to Enable Highly Localized Computing-On-the-Move DNN Dataflow, IEEE Transactions on Circuits and Systems II: Express Briefs. (2021).
[6] Y. Wang, H. Li, Y. Han & X. Li, A Low Overhead In-Network Data Compressor for the Memory Hierarchy of Chip Multiprocessors, IEEE Trans. Comput.-Aided Des. Integr. Circuits Syst. 37(6) (2018) 1265–1277.
[7] S. Kim & Y. Kim, Novel XNOR-based Approximate Computing for Energy-Efficient Image Processors, Journal of Semiconductor Technology and Science. 18(5) (2018) 602-608.
[8] Y. He, X. Yi, Z. Zhang, B. Ma & Q. Li, A Probabilistic Prediction-Based Fixed-Width Booth Multiplier for Approximate Computing, IEEE Transactions on Circuits and Systems I: Regular Papers. 67(12) (2020) 4794-4803.
[9] F. Betzel, K. Khatamifard, H. Suresh, D.J. Lilja, J. Sartori & U. Karpuzcu, Approximate Communication: Techniques for Reducing Communication Bottlenecks in Large-Scale Parallel Systems, ACM Comput, Surveys. 51(1) (2018) 1–32.
[10] S. Xiao, X. Wang, M. Palesi, A.K. Singh & T. Mak, ACDC: An Accuracy- And Congestion-Aware Dynamic Traffic Control Method for Networks-on-Chip, In Proc. IEEE Design Autom. Test Eur. Conf. Exhibit, (DATE), Florence, Italy. (2019) 630–633.
[11] R. Fenster & S. Le Beux, RELAX: A Reconfigurable Approximate Network-on-Chip, In IEEE 14th International Symposium on Embedded Multicore/Many-core Systems-on-Chip (MCSoC), IEEE. (2021) 381-387.
[12] Y. Chen & A. Louri, Learning-Based Quality Management for Approximate Communication in Network-on-Chips, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. 39(11) (2020) 3724-3735.
[13] A. Jain, P. Hill, S.C. Lin, M. Khan, M.E. Haque, M.A. Laurenzano & J. Mars, Concise Loads and Stores: The Case for an Asymmetric Compute-Memory Architecture for Approximation, In Proc. of MICRO. (2016).
[14] A. Ranjan, A. Raha, V. Raghunathan & A. Raghunathan, Approximate Memory Compression for Energy-Efficiency, In Proc. of ISLPED. (2017).
[15] J.S. Miguel, J. Albericio, A. Moshovos & N.E. Jerger, Doppelgänger: A Cache for Approximate Computing, In Proc. of MICRO. (2015).
[16] R. Boyapati, J. Huang, P. Majumder, K.H. Yum & E.J. Kim, APPROXNoC: A Data Approximation Framework for Network-On-Chip Architectures, In Proc. of ISCA. (2017).
[17] Q. Wang, Y. Li & P. Li, Liquid State Machine Based Pattern Recognition on FPGA with Firing-Activity Dependent Power Gating and Approximate Computing, In 2016 IEEE International Symposium on Circuits and Systems ISCAS, IEEE. (2016) 361-364.
[18] J.S. Kim, J.B. Hong, J.Y. Kang & T.H. Han, Lifetime Improvement Method Using Threshold-Based Partial Data Compression in Noc, In 2018 International SoC Design Conference ISOCC, IEEE. (2018) 269-270.
[19] N. Niwa, Y. Shikama, H. Amano & M. Koibuchi, A Case for Low-Latency Network-on-Chip Using Compression Routers, In 2021 29th Euromicro International Conference on Parallel, Distributed and Network-Based Processing (PDP), IEEE. (2021) 134-142.
[20] J. Zhan, M. Poremba, Y. Xu & Y. Xie, NoD: Leveraging Delta Compression for End-to-End Memory Access in Noc Based Multicores, in Proc. 19th Asia South Pacific Des. Automat. Conf. (2014) 586–591
[21] Y. Zhang, Y. Yuan, D. Feng, C. Wang, X. Wu, L. Yan & S. Wang, Improving Restore Performance for In-Line Backup System Combining Deduplication and Delta Compression, IEEE Trans. Parallel Distrib. Syst. 31(10) (2020) 2302–2314.
[22] J.R. Stevens, A. Ranjan, & A. Raghunathan, AxBA: An Approximate Bus Architecture Framework, In Proceedings of the International Conference on Computer-Aided Design. (2018) 1-8.
[23] Y. Chen & A. Louri, An Online Quality Management Framework for Approximate Communication in Network-On-Chips, In Proceedings of the ACM International Conference on Supercomputing. (2019) 217-226.
[24] Y. Chen, M.F. Reza & A. Louri, DEC-NoC: An Approximate Framework Based on Dynamic Error Control with Applications to Energy-Efficient NoCs, In 2018 IEEE 36th International Conference on Computer Design ICCD, IEEE. (2018) 480-487.
[25] S. Xiao, X. Wang, M. Palesi, A.K. Singh, L. Wang, & T. Mak, On Performance Optimization and Quality Control for Approximate-Communication-Enabled Networks-on-Chip, IEEE Transactions on Computers. 70(11) (2020) 1817-1830.
[26] Y. Chen & A. Louri, An Approximate Communication Framework for Network-on-Chips, IEEE Transactions on Parallel and Distributed Systems. 31(6) (2020) 1434-1446.
[27] IEEE Standards Committee et al., IEEE Standard for Floating-Point Arithmetic, IEEE Computer Society Standard. 517 (2008).
[28] Krutthika H.K, A R Aswatha, Design of Efficient FSM Based 3D Network on Chip Architecture, International Journal of Engineering Trends and Technology. 68(10) (2020):67-73.