針對三維多核處理器效能優化之一種以溫度為考量的任務分配方法

A Thermal-Driven Task Allocation for 3D MCP Throughput Optimization

何宜倫(Yi-Lun Ho)

指導教授 ： 張世杰

國立清華大學/電機資訊學院/資訊工程學系/碩士(2010年)

https://doi.org/10.6843/NTHU.2010.00459

摘要

三維晶片已經變成一個新興的技術，由於它可以提高封裝密度以及相當大的異質整合彈性。然而在三維晶片中有著不易散熱的問題，使得在考慮優化三維多核處理器效能時，需要把高溫的影響考慮進來。此外在三維多何處理器裡，溫度受任務位置分配的影響很大，一個好的任務分配可以減少許多溫度帶來的問題以及增加效能。另外在這篇論文中，為了達到連線排程，我們提出了一個非常高效率的任務分配演算法可以在溫度限制下優化三維多核處理器效能。實驗結果顯示我們快速的演算法可以使128的多核處理器下的任務分配，在1微秒內完成。並且與LINGO軟體結果比較，我們的方法平均只差0.85%的效能。

關鍵字

三維晶片 ； 效率優化 ； 任務分配

並列摘要

Three-dimensional integrated circuits (3D ICs) have become an emerging technology because of advantages of higher packaging density and more flexible in heterogeneous integration. However, due to the heat dissipation problem in 3D ICs, thermal issues must be considered when maximizing the throughput of 3D Multi-Core Processors (MCPs). Furthermore, since the thermal behavior of a core in 3D MCPs strongly depends on its location, a proper task allocation can reduce potential thermal problems and improve the throughput. In this thesis, to enable the possibility of on-line scheduling, we present a very high performance task allocation algorithm to solve the problem of throughput optimization under thermal constraints for 3D MCPs. The experimental results show our fast algorithm can finish in less than 1ms for 128 cores. Compared to solutions obtained from LINGO, our algorithm only has 0.85% throughput loss on average.

並列關鍵字

3D Multi-Core Processor ； Throughput optimization ； Task allocation

參考文獻

[2]V. Hanumaiah, R. Rao, S. Vrudhula and K. S. Chatha, “Throughput Optimal Task Allocation under Thermal Constraints for Multi-core Processors,” in Proc. of DAC, pp. 776-781, 2009

[4]W. Huang, K. Skadron, S. Gurumurthi, R. J. Ribando, and M. R. Stan, “Differentiating the Roles of IR Measurement and Simulation for Power and Temperature-Aware Design,” in Proceedings of the 2009 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS), Apr. 2009.

[5]M. Kadin and S. Reda, “Frequency and Voltage Planning for Multi-Core Processors Under Thermal Constraints,” in IEEE International Conference on Computer Design, pp. 463-470, 2008.

[7]W. Liao, L. He, and K. Lepak, “Temperature and Supply Voltage Aware Performance and Power Modeling at Microarchitecture Level,” Transactions on Computer-Aided Design of Integrated Circuits and Systems, vol. 24(7), pp. 1042--1053, 2005.

[8]R. Rao and S. Vrudhula. “Efficient online computation of core speeds to maximize the throughput of thermally constrained multi-core processors,” in Proc. of ICCAD, pp. 537-542, 2008.