Leakage power characterization and minimization in 3D stacked multi-core chips with microfluidic cooling

The needs of multiple-functionality and low cost have driven the development of high-density electronic packages. However, the greater package density results in higher power density per unit volume of the package, which creates challenges for thermal management. Microfluidic cooling can potentially achieve superior thermal performance with surface area enhancements such as pin fins and could be a viable solution for many applications to the increasing power density in electronic packages. In this paper, we report on investigations of the impact of the microfluidic cooling technology on the system level performance of multicore architectures stacked in a 3D package. Specifically, we characterize the impact on leakage power dissipation over different pin fin configurations and its impact on overall system energy efficiency. We do so with a cycle-level application-driven full system simulation framework. The framework executes application binaries and operating system code and models coupled interactions among the i) application & operating system code, ii) resulting thermal field, iii) leakage power, and v) microfluidic cooling. This provides the unique ability to assess the impact of microfluidics on computing system level metrics experienced by the applications such as energy per instruction. We illustrate and quantify improvements in energy efficiency of the applications, as well as increase in throughput due to microfluidic cooling.

[1]  Sudhakar Yalamanchili,et al.  Co-design of multicore architectures and microfluidic cooling for 3D stacked ICs , 2013 .

[2]  Jung Ho Ahn,et al.  McPAT: An integrated power, area, and timing modeling framework for multicore and manycore architectures , 2009, 2009 42nd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO).

[3]  David Atienza,et al.  3D-ICE: Fast compact transient thermal modeling for 3D ICs with inter-tier liquid cooling , 2010, 2010 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).

[4]  M. Szermer,et al.  Cache leakage power estimation using architectural model for 32 nm and 16 nm technology nodes , 2012, 2012 28th Annual IEEE Semiconductor Thermal Measurement and Management Symposium (SEMI-THERM).

[5]  J. Jeddeloh,et al.  Hybrid memory cube new DRAM architecture increases density and performance , 2012, 2012 Symposium on VLSI Technology (VLSIT).

[6]  Jun Wang,et al.  Manifold: A parallel simulation framework for multicore systems , 2014, 2014 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS).

[7]  Kai Li,et al.  PARSEC vs. SPLASH-2: A quantitative comparison of two multithreaded benchmark suites on Chip-Multiprocessors , 2008, 2008 IEEE International Symposium on Workload Characterization.

[8]  Soha Hassoun,et al.  System-level comparison of power delivery design for 2D and 3D ICs , 2009, 2009 IEEE International Conference on 3D System Integration.

[9]  Yogendra Joshi,et al.  Compact Modeling of 3D Stacked Die Inter-Tier Microfluidic Cooling Under Non-Uniform Heat Flux , 2012 .

[10]  S. Gurrum,et al.  Thermal Issues in Next Generation Integrated Circuits , 2003 .

[11]  Chia-Jung Hsu Numerical Heat Transfer and Fluid Flow , 1981 .