Heat-and-run: leveraging SMT and CMP to manage power density through the operating system
ASPLOS XI Proceedings of the 11th international conference on Architectural support for programming languages and operating systems
Thermal-Aware Clustered Microarchitectures
ICCD '04 Proceedings of the IEEE International Conference on Computer Design
Scheduling for heterogeneous processors in server systems
Proceedings of the 2nd conference on Computing frontiers
Variability and energy awareness: a microarchitecture-level perspective
Proceedings of the 42nd annual Design Automation Conference
A Linux-based tool for hardware bring up, Linux development, and manufacturing
IBM Systems Journal
Techniques for Multicore Thermal Management: Classification and New Exploration
Proceedings of the 33rd annual international symposium on Computer Architecture
Power efficiency for variation-tolerant multicore processors
Proceedings of the 2006 international symposium on Low power electronics and design
High-performance CMOS variability in the 65-nm regime and beyond
IBM Journal of Research and Development - Advanced silicon technology
Microarchitecture parameter selection to optimize system performance under process variation
Proceedings of the 2006 IEEE/ACM international conference on Computer-aided design
ReCycle:: pipeline adaptation to tolerate process variation
Proceedings of the 34th annual international symposium on Computer architecture
Impact of process variations on multicore performance symmetry
Proceedings of the conference on Design, automation and test in Europe
Thermal-aware task scheduling at the system software level
ISLPED '07 Proceedings of the 2007 international symposium on Low power electronics and design
Thermal-aware scheduling for future chip multiprocessors
EURASIP Journal on Embedded Systems
Mitigating Parameter Variation with Dynamic Fine-Grain Body Biasing
Proceedings of the 40th Annual IEEE/ACM International Symposium on Microarchitecture
Proceedings of the 5th conference on Computing frontiers
Addressing thermal nonuniformity in SMT workloads
ACM Transactions on Architecture and Code Optimization (TACO)
Variation-Aware Application Scheduling and Power Management for Chip Multiprocessors
ISCA '08 Proceedings of the 35th Annual International Symposium on Computer Architecture
Many-core design from a thermal perspective
Proceedings of the 45th annual Design Automation Conference
ACM Transactions on Design Automation of Electronic Systems (TODAES)
Power/Performance/Thermal Design-Space Exploration for Multicore Architectures
IEEE Transactions on Parallel and Distributed Systems
Static and dynamic temperature-aware scheduling for multiprocessor SoCs
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
EVAL: Utilizing processors with variation-induced timing errors
Proceedings of the 41st annual IEEE/ACM International Symposium on Microarchitecture
Mitigating the impact of variability on chip-multiprocessor power and performance
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Exploiting unbalanced thread scheduling for energy and performance on a CMP of SMT processors
IPDPS'06 Proceedings of the 20th international conference on Parallel and distributed processing
A novel software solution for localized thermal problems
ISPA'06 Proceedings of the 4th international conference on Parallel and Distributed Processing and Applications
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Understanding the Thermal Implications of Multi-Core Architectures
IEEE Transactions on Parallel and Distributed Systems
Proceedings of the 49th Annual Design Automation Conference
Spatial and temporal thermal characterization of stacked multicore architectures
ACM Journal on Emerging Technologies in Computing Systems (JETC)
Hi-index | 0.00 |
Inherent temperature variation among cores in a multicore architecture can be caused by a number of factors including process variation, cooling and packaging imperfections, and even placement of the chip in the module. Current dynamic thermal management techniques assume identical heating profiles for homogeneous multicore architectures. Our experimental results indicate that inherent thermal variation is very common in existing multicores. While most multicore chips accommodate multiple thermal sensors, the dynamic power/thermal management schemes are oblivious of the inherent heating tendencies. Hence, in the case of variation, the chip faces repetitive hotspots running on such cores. In this article, we propose a technique that leverages the on-chip sensor infrastructure as well as the capabilities of power/thermal management to effectively reduce the heating and minimize local hotspots. This technique can be used in existing multicore chips as long as the thermal sensor data can be made transparent to the power/thermal management at the software layer. According to our experimental analysis on test-chips, 5°C peak temperature reduction can be achieved with no performance degradation, hence the inherent energy efficiency of the chip can be improved without any performance or cost penalty.