Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Thermal-aware Adaptive Platform Management for Heterogeneous Embedded Systems

Thermal-aware Adaptive Platform Management for Heterogeneous Embedded Systems Recent trends in real-time applications have raised the demand for high-throughput embedded platforms with integrated CPU-GPU based Systems-On-Chip (SoCs). The enhanced performance of such SoCs, however, comes at the cost of increased power consumption, resulting in significant heat dissipation and high on-chip temperatures. The prolonged occurrences of high on-chip temperature can cause accelerated in-circuit ageing, which severely degrades the long-term performance and reliability of the chip. Violation of thermal constraints leads to on-board dynamic thermal management kicking-in, which may result in timing unpredictability for real-time tasks due to transient performance degradation. Recent work in adaptive software design have explored this issue from a control theoretic stand-point, striving for smooth thermal envelopes by tuning the core frequency.Existing techniques do not handle thermal violations for periodic real-time task sets in the presence of dynamic events like change of task periodicity, more so in the context of heterogeneous SoCs with integrated CPU-GPUs. This work presents an OpenCL runtime extension for thermal-aware scheduling of periodic, real-time tasks on heterogeneous multi-core platforms. Our framework mitigates dynamic thermal violations by adaptively tuning task mapping parameters, with the eventual control objective of satisfying both platform-level thermal constraints and task-level deadline constraints. We consider multiple platform-level control actions like task migration, frequency tuning and idle slot insertion as the task mapping parameters. To the best of our knowledge, this is the first work that considers such a variety of task mapping control actions in the context of heterogeneous embedded platforms. We evaluate the proposed framework on an Odroid-XU4 board using OpenCL benchmarks and demonstrate its effectiveness in reducing thermal violations. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png ACM Transactions on Embedded Computing Systems (TECS) Association for Computing Machinery

Thermal-aware Adaptive Platform Management for Heterogeneous Embedded Systems

Download PDF
28 pages

Loading next page...
 
/lp/association-for-computing-machinery/thermal-aware-adaptive-platform-management-for-heterogeneous-embedded-yLLTxqkue6

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Association for Computing Machinery
Copyright
Copyright © 2021 Association for Computing Machinery.
ISSN
1539-9087
eISSN
1558-3465
DOI
10.1145/3477028
Publisher site
See Article on Publisher Site

Abstract

Recent trends in real-time applications have raised the demand for high-throughput embedded platforms with integrated CPU-GPU based Systems-On-Chip (SoCs). The enhanced performance of such SoCs, however, comes at the cost of increased power consumption, resulting in significant heat dissipation and high on-chip temperatures. The prolonged occurrences of high on-chip temperature can cause accelerated in-circuit ageing, which severely degrades the long-term performance and reliability of the chip. Violation of thermal constraints leads to on-board dynamic thermal management kicking-in, which may result in timing unpredictability for real-time tasks due to transient performance degradation. Recent work in adaptive software design have explored this issue from a control theoretic stand-point, striving for smooth thermal envelopes by tuning the core frequency.Existing techniques do not handle thermal violations for periodic real-time task sets in the presence of dynamic events like change of task periodicity, more so in the context of heterogeneous SoCs with integrated CPU-GPUs. This work presents an OpenCL runtime extension for thermal-aware scheduling of periodic, real-time tasks on heterogeneous multi-core platforms. Our framework mitigates dynamic thermal violations by adaptively tuning task mapping parameters, with the eventual control objective of satisfying both platform-level thermal constraints and task-level deadline constraints. We consider multiple platform-level control actions like task migration, frequency tuning and idle slot insertion as the task mapping parameters. To the best of our knowledge, this is the first work that considers such a variety of task mapping control actions in the context of heterogeneous embedded platforms. We evaluate the proposed framework on an Odroid-XU4 board using OpenCL benchmarks and demonstrate its effectiveness in reducing thermal violations.

Journal

ACM Transactions on Embedded Computing Systems (TECS) – Association for Computing Machinery

Published: Sep 22, 2021

Keywords: Heterogeneous computing

References

  • Feedback performance control in software services
    Abdelzaher, Tarek F.; Stankovic, John A.; Lu, Chenyang; Zhang, Ronghua; Lu, Ying
  • On the efficiency of voltage overscaling under temperature and aging effects
    Amrouch, Hussam; Ehsani, Seyed Borna; Gerstlauer, Andreas; Henkel, Jörg
  • Coordinated management of multiple interacting resources in chip multiprocessors: A machine learning approach
    Bitirgen, Ramazan; Ipek, Engin; Martinez, Jose F.
  • Dynamic thermal management for high-performance microprocessors
    Brooks, David; Martonosi, Margaret
  • Temperature-aware scheduling and assignment for hard real-time applications on MPSoCs
    Chantem, Thidapat; Hu, X. Sharon; Dick, Robert P.
  • A thermal-aware scheduling for multicore architectures
    Chien, Ting-Hsuan; Chang, Rong-Guey
  • Bin packing approximation algorithms: Combinatorial analysis
    Coffman, Edward G.; Galambos, Gabor; Martello, Silvano; Vigo, Daniele
  • Paragon: QoS-aware scheduling for heterogeneous datacenters
    Delimitrou, Christina; Kozyrakis, Christos
  • Coscale: Coordinating cpu and memory system dvfs in server systems
    Deng, Qingyuan; Meisner, David; Bhattacharjee, Abhishek; Wenisch, Thomas F.; Bianchini, Ricardo
  • Scheduling challenges and opportunities in integrated CPU+GPU processors
    Dev, Kapil; Reda, Sherief
  • EdgeCoolingMode: An Agent based Thermal Management Mechanism for DVFS enabled Heterogeneous MPSoCs
    Dey, Somdip; Guajardo, Enrique Zaragoza; Basireddy, Karunakar Reddy; Wang, Xiaohang; Singh, Amit Kumar; McDonald-Maier, Klaus
  • Managing multi-configuration hardware via dynamic working set analysis
    Dhodapkar, Ashutosh S.; Smith, James E.
  • Distributed task migration for thermal management in many-core systems
    Ge, Yang; Malani, Parth; Qiu, Qinru
  • Divergence aware automated partitioning of OpenCL workloads
    Ghose, Anirban; Dey, Soumyajit; Mitra, Pabitra; Chaudhuri, Mainak
  • A framework for OpenCL task scheduling on heterogeneous multicores
    Ghose, Anirban; Dokara, Lokesh; Dey, Soumyajit; Mitra, Pabitra
  • Orchestration of perception systems for reliable performance in heterogeneous platforms
    Ghose, Anirban; Maity, Srijeeta; Kar, Arijit; Dey, Soumyajit
  • A survey of physics-based attack detection in cyber-physical systems
    Giraldo, Jairo; Urbina, David; Cardenas, Alvaro; et al,
  • Multirate controller design for resource-and schedule-constrained automotive ECUs
    Goswami, Dip; Masrur, Alejandro; Schneider, Reinhard; Xue, Chun Jason; Chakraborty, Samarjit
  • Auto-tuning a high-level language targeted to GPU codes
    Grauer-Gray, Scott; Xu, Lifan; Searles, Robert; Ayalasomayajula, Sudhee; Cavazos, John
  • A static task partitioning approach for heterogeneous systems using OpenCL
    Grewe, Dominik; O’Boyle, Michael F. P.
  • Thermal response to DVFS: Analysis with an Intel Pentium M
    Hanson, Heather; Keckler, Stephen W.; Ghiasi, Soraya; Rajamani, Karthick; Rawson, Freeman; Rubio, Juan
  • Feedback control of computing systems
    Hellerstein, Joseph L.; Diao, Yixin; Parekh, Sujay; Tilbury, Dawn M.
  • POET: A portable approach to minimizing energy under soft real-time constraints
    Imes, Connor; Kim, David H. K.; Maggio, Martina; Hoffmann, Henry
  • Self-optimizing memory controllers: A reinforcement learning approach
    Ipek, Engin; Mutlu, Onur; Martínez, José F.; Caruana, Rich
  • TEEM: Online thermal-and energy-efficiency management on CPU-GPU MPSoCs
    Isuwa, Samuel; Dey, Somdip; Singh, Amit Kumar; McDonald-Maier, Klaus
  • M-DTM: Migration-based dynamic thermal management for heterogeneous mobile multi-core processors
    Kim, Young Geun; Kim, Minyong; Kim, Jae Min; Chung, Sung Woo
  • An automatic input-sensitive approach for heterogeneous task partitioning
    Kofler, Klaus; Grasso, Ivan; Cosenza, Biagio; Fahringer, Thomas
  • Keep it slow and in time: Online DVFS with hard real-time workloads
    Lampka, Kai; Forsberg, Björn
  • Transparent CPU-GPU collaboration for data-parallel kernels on heterogeneous systems
    Lee, Janghaeng; Samadi, Mehrzad; Park, Yongjun; Mahlke, Scott
  • Thermal-aware resource management for embedded real-time systems
    Lee, Youngmoon; Chwa, Hoon Sung; Shin, Kang G.; Wang, Shige
  • Thermal-aware scheduling for integrated CPUs-GPU platforms
    Lee, Youngmoon; Shin, Kang G.; Chwa, Hoon Sung
  • Feedback control real-time scheduling: Framework, modeling, and algorithms
    Lu, Chenyang; Stankovic, John A.; Son, Sang H.; Tao, Gang
  • Thermal load-aware adaptive scheduling for heterogeneous platforms
    Maity, Srijeeta; Ghose, Anirban; Dey, Soumyajit; Biswas, Swarnendu
  • Dynamic GPGPU power management using adaptive model predictive control
    Majumdar, Abhinandan; Piga, Leonardo; Paul, Indrani; Greathouse, Joseph L.; Huang, Wei; Albonesi, David H.
  • CALOREE: Learning control for predictable latency and low energy
    Mishra, Nikita; Imes, Connor; Lafferty, John D.; Hoffmann, Henry
  • Mixed-criticality real-time scheduling for multicore systems
    Mollison, Malcolm S.; Erickson, Jeremy P.; Anderson, James H.; Baruah, Sanjoy K.; Scoredos, John A.
  • The opencl specification
    Munshi, Aaftab
  • Improving mobile gaming performance through cooperative CPU-GPU thermal management
    Prakash, Alok; Amrouch, Hussam; Shafique, Muhammad; Mitra, Tulika; Henkel, Jörg
  • SPECTR: Formal supervisory control and coordination for many-core systems resource management
    Rahmani, Amir M.; Donyanavard, Bryan; Mück, Tiago; et al,
  • Optimum: Thermal-aware task allocation for heterogeneous many-core devices
    Rudi, Andrea; Bartolini, Andrea; Lodi, Andrea; Benini, Luca
  • Traffic sign recognition with multi-scale Convolutional Networks
    Sermanet, Pierre; LeCun, Yann
  • Predictive dynamic thermal and power management for heterogeneous mobile platforms
    Singla, Gaurav; Kaur, Gurinderjit; Unver, Ali K.; Ogras, Umit Y.
  • Predictive dynamic thermal and power management for heterogeneous mobile platforms
    Singla, Gaurav; Kaur, Gurinderjit; Unver, Ali K.; Ogras, Umit Y.
  • Hybrid architectural dynamic thermal management
    Skadron, Kevin
  • Quality-adaptive deep learning for pedestrian detection
    Tahboub, Khalid; Güera, David; Reibman, Amy R.; Delp, Edward J.
  • OPTiC: Optimizing collaborative CPU–GPU computing on mobile devices with thermal constraints
    Wang, Siqi; Ananthanarayanan, Gayathri; Mitra, Tulika
  • Cache contention and application performance prediction for multi-core systems
    Xu, Chi; Chen, Xi; Dick, Robert P.; Mao, Zhuoqing Morley
  • Robust multi-resolution pedestrian detection in traffic scenes
    Yan, Junjie; Zhang, Xucong; Lei, Zhen; Liao, Shengcai; Li, Stan Z.
  • Maximizing performance under a power cap: A comparison of hardware, software, and hybrid techniques
    Zhang, Huazhe; Hoffmann, Henry
  • Co-run scheduling with power cap on integrated CPU-GPU systems
    Zhu, Qi; Wu, Bo; Shen, Xipeng; Shen, Li; Wang, Zhiying