Abstract
Many-Core Systems-on-Chip increasingly require Dynamic Multi-objective Management (DMOM) of resources. DMOM uses different management components for objectives and resources to implement comprehensive and self-adaptive system resource management. DMOMs are challenging because they require a scalable and well-organized framework to make each component modular, allowing it to be instantiated or redesigned with a limited impact on other components.
This work evaluates two state-of-the-art distributed management paradigms and, motivated by their drawbacks, proposes a new one called Management Application (MA), along with a DMOM framework based on MA. MA is a distributed application, specific for management, where each task implements a management role. This paradigm favors scalability and modularity because the management design assumes different and parallel modules, decoupled from the OS.
An experiment with a task mapping case study shows that MA reduces the overhead of management resources (-61.5%), latency (-66%), and communication volume (-96%) compared to state-of-the-art per-application management. Compared to cluster-based management (CBM) implemented directly as part of the OS, MA is similar in resources and communication volume, increasing only the mapping latency (+16%). Results targeting a complete DMOM control loop addressing up to three different objectives show the scalability regarding system size and adaptation frequency compared to CBM, presenting an overall management latency reduction of 17.2% and an overall monitoring messages’ latency reduction of 90.2%.
- M. Al Faruque, J. Jahn, T. Ebi, and J. Henkel. 2010. Runtime thermal management using software agents for multi- and many-core architectures. IEEE Design Test of Computers 27, 6 (2010), 58–68. DOI:https://doi.org/10.1109/MDT.2010.94Google Scholar
Digital Library
- I. Anagnostopoulos, V. Tsoutsouras, A. Bartzas, and D. Soudris. 2013. Distributed run-time resource management for malleable applications on many-core platforms. In DAC. ACM, 168:1–168:6. DOI:https://doi.org/10.1145/2463209.2488942Google Scholar
- B. Bohnenstiehl, A. Stillmaker, J. Pimentel, T. Andreas, B. Liu, A. Tran, E. Adeagbo, and B. Baas. 2017. KiloCore: A 32-nm 1000-processor computational array. J. Solid-State Circuits 52, 4 (2017), 891–902. DOI:https://doi.org/10.1109/JSSC.2016.2638459Google Scholar
Cross Ref
- E. Carara, N. Calazans, and F. G. Moraes. 2014. Differentiated communication services for NoC-Based MPSoCs. IEEE Trans. Computers 63, 3 (2014), 595–608. DOI:https://doi.org/10.1109/TC.2012.123Google Scholar
Digital Library
- B. D. de Dinechin, R. Ayrignac, P. Beaucamps, P. Couvert, B. Ganne, P. G. de Massas, F. Jacquet, S. Jones, N. M. Chaisemartin, F. Riss, and T. Strudel. 2013. A clustered manycore processor architecture for embedded and accelerated applications. In HPEC. IEEE, 1–6. DOI:https://doi.org/10.1109/HPEC.2013.6670342Google Scholar
- M. Fattah, M. Daneshtalab, P. Liljeberg, and J. Plosila. 2011. Exploration of MPSoC monitoring and management systems. In ReCoSoC. IEEE, 1–3. DOI:https://doi.org/10.1109/ReCoSoC.2011.5981544Google Scholar
- H. Hoffmann, J. Eastep, M. D. Santambrogio, J. E. Miller, and A Agarwal. 2010. Application heartbeats: A generic interface for specifying program performance and goals in autonomous computing environments. In ICAC. ACM, 79–88. DOI:https://doi.org/10.1145/1809049.1809065Google Scholar
- H. Khdr, S. Pagani, M. Shafique, and J. Henkel. 2018. Chapter four - dark silicon aware resource management for many-core systems. In Dark Silicon and Future On-chip Systems, Ali R. Hurson and Hamid Sarbazi-Azad (Eds.). Elsevier, 127–170. DOI:https://doi.org/10.1016/bs.adcom.2018.03.002Google Scholar
- S. Kobbe, L. Bauer, D. Lohmann, W. Schröder-Preikschat, and J. Henkel. 2011. DistRM: Distributed resource management for on-chip many-core systems. In CODES+ISSS. ACM, 119–128. DOI:https://doi.org/10.1145/2039370.2039392Google Scholar
- A. Martins, A. L. da Silva, A. Rahmani, N. Dutt, and F. G. Moraes. 2019. Hierarchical adaptive multi-objective resource management for many-core systems. Journal of Systems Architecture 97 (2019), 416–427. DOI:https://doi.org/10.1016/j.sysarc.2019.01.006Google Scholar
Digital Library
- A. Miele, A. Kanduri, K. Moazzemi, D. Juhász, A. Rahmani, N. D. Dutt, P. Liljeberg, and A. Jantsch. 2019. On-Chip dynamic resource management. Foundations and Trends in Electronic Design Automation 13, 1-2 (2019), 1–14. DOI:https://doi.org/10.1561/1000000055Google Scholar
Digital Library
- S. Pagani, H. Khdr, J. Chen, M. Shafique, M. Li, and J. Henkel. 2017. Thermal safe power (TSP): Efficient power budgeting for heterogeneous manycore systems in dark silicon. IEEE Trans. Comput. 66, 1 (2017), 147–162. DOI:https://doi.org/10.1109/TC.2016.2564969Google Scholar
Digital Library
- Wei Quan and Andy D Pimentel. 2016. A hierarchical run-time adaptive resource allocation framework for large-scale MPSoC systems. Design Automation for Embedded Systems 20, 4 (2016), 311–339. DOI:https://doi.org/10.1007/s10617-016-9179-zGoogle Scholar
Digital Library
- S. Rhoads. 2016. Plasma - most MIPS I(TM). https://opencores.org/projects/plasma.Google Scholar
- M. Ruaro, L. Caimi, V. Fochi, and F. G. Moraes. 2019. Memphis: A framework for heterogeneous many-core SoCs generation and validation. Design Automation for Embedded Systems 23, 3 (2019), 103–122. DOI:https://doi.org/10.1007/s10617-019-09223-4Google Scholar
Cross Ref
- M. Ruaro, E. Carara, and F. Moraes. 2015. Runtime adaptive circuit switching and flow priority in NoC-Based MPSoCs. IEEE Trans. Very Large Scale Integr. Syst 23, 6 (2015), 1077–1088. DOI:https://doi.org/10.1109/TVLSI.2014.2331135Google Scholar
Digital Library
- M. Ruaro, A. Jantsch, and F. G. Moraes. 2019. Self-Adaptive QoS management of computation and communication resources in many-core SoCs. ACM Transaction on Embedded Computing Systems 18, 4 (2019), 37:1–37:21. DOI:https://doi.org/10.1145/3328755Google Scholar
- M. Ruaro and F. Moraes. 2017. Demystifying the cost of task migration in distributed memory many-core systems. In ISCAS. IEEE, 1–4. DOI:https://doi.org/10.1109/ISCAS.2017.8050257Google Scholar
- S. Saponara, T. Bacchillone, E. Petri, L. Fanucci, R. Locatelli, and M. Coppola. 2014. Design of an NoC interface macrocell with hardware support of advanced networking functionalities. IEEE Trans. Comput. 63, 03 (2014), 609–621. DOI:https://doi.org/10.1109/TC.2012.70Google Scholar
Digital Library
- S. Sarma, N. D. Dutt, P. Gupta, A. Nicolau, and N. Venkatasubramanian. 2014. On-chip self-awareness using Cyberphysical-Systems-on-Chip (CPSoC). In CODES+ISSS. ACM, 22:1–22:3. DOI:https://doi.org/10.1145/2656075.2661648Google Scholar
- M. Shafique and S. Garg. 2017. Computing in the dark silicon era: Current trends and research challenges. IEEE Design 8 Test 34, 2 (2017), 8–23. DOI:https://doi.org/10.1109/MDAT.2016.2633408Google Scholar
- M. Shafique, B. Vogel, and J. Henkel. 2013. Self-adaptive hybrid dynamic power management for many-core systems. In DATE. ACM, 51–56. DOI:https://doi.org/10.7873/DATE.2013.025Google Scholar
- E. Shamsa, A. Kanduri, A. M. Rahmani, P. Liljeberg, A. Jantsch, and N. Dutt. 2019. Goal-Driven autonomy for efficient on-chip resource management: transforming objectives to goals. In DATE. IEEE, 1397–1402. DOI:https://doi.org/10.23919/DATE.2019.8715134Google Scholar
- A. K. Singh, M. Shafique, A. Kumar, and J. Henkel. 2013. Mapping on multi/many-core systems: Survey of current and emerging trends. In DAC. ACM, 1–10. DOI:https://doi.org/10.1145/2463209.2488734Google Scholar
- V. Tsoutsouras, I. Anagnostopoulos, D. Masouros, and D. Soudris. 2018. A hierarchical distributed runtime resource management scheme for NoC-Based many-cores. ACM Transactions on Embedded Computing Systems 17, 3 (2018), 65:1–65:26. DOI:https://doi.org/10.1145/3182173Google Scholar
Digital Library
- S. Wildermann, M. Glaß, and J. Teich. 2014. Multi-objective distributed run-time resource management for many-cores. In DATE. IEEE, 1–6. DOI:https://doi.org/10.7873/DATE.2014.234Google Scholar
- Y. Xiao, S. Nazarian, and P. Bogdan. 2019. Self-Optimizing and self-programming computing systems: A combined compiler, complex networks, and machine learning approach. IEEE Transactions on Very Large Scale Integration (VLSI) Systems 27, 6 (2019), 1416–1427. DOI:https://doi.org/10.1109/TVLSI.2019.2897650Google Scholar
Digital Library
- Y. Xue, Z. Qian, G. Wei, P. Bogdan, C. Tsui, and R. Marculescu. 2014. An efficient Network-on-Chip (NoC) based multicore platform for hierarchical parallel genetic algorithms. In NOCS. ACM, 17–24. DOI:https://doi.org/10.1109/NOCS.2014.7008757Google Scholar
- H. Zhang and H. Hoffmann. 2016. Maximizing performance under a power cap: A comparison of hardware, software, and hybrid techniques. In ASPLOS. ACM, 545–559. DOI:https://doi.org/10.1145/2872362.2872375Google Scholar
Index Terms
Modular and Distributed Management of Many-Core SoCs
Recommendations
Implementation and Evaluation of Large Interconnection Routers for Future Many-core Networks on Chip
HPCC '12: Proceedings of the 2012 IEEE 14th International Conference on High Performance Computing and Communication & 2012 IEEE 9th International Conference on Embedded Software and SystemsAs the number of processing elements in the future Networks on Chip (NoC) increases from multi-cores to many-cores, the role of the interconnection communications becomes more critical. The number of cores on a System on Chip (SoC) will reach thousands ...
Distributed SDN architecture for NoC-based many-core SoCs
NOCS '19: Proceedings of the 13th IEEE/ACM International Symposium on Networks-on-ChipIn the Software-Defined Networking (SDN) paradigm, routers are generic and programmable forwarding units that transmit packets according to a given policy defined by a software controller. Recent research has shown the potential of such a communication ...
Distributed sensor data processing for many-cores
GLSVLSI '12: Proceedings of the great lakes symposium on VLSIFuture many-core systems will rely heavily on a wide variety of sensors which provide run-time information about on-chip environment and workload. In this paper, a new dedicated infrastructure for distributed sensor processing for many-core systems is ...






Comments