The MTA: An Advanced and Versatile Thermal Simulator for Integrated Systems

Abstract

Fast and accurate thermal analysis is crucial for determining the propagation of heat and tracking the formation of hot spots in integrated circuits (ICs). Existing academic thermal analysis tools primarily use compact models to accelerate thermal simulations but are limited to linear problems on relatively simple circuit geometries. The Manchester Thermal Analyzer (MTA) is a comprehensive tool that allows for fast and highly accurate linear and nonlinear thermal simulations of complex physical structures including the IC, the package, and the heatsink. The MTA is targeted for 2.5/3-D IC designs but also handles standard planar ICs. The MTA discretizes the heat equation in space using the finite element method and performs the time integration with unconditionally stable implicit time stepping methods. To improve the computational efficiency without sacrificing accuracy, the MTA features adaptive spatiotemporal refinement. The largescale linear systems that arise during the simulation are solved with fast preconditioned Krylov subspace methods. The MTA supports the thermal analysis of realistic integrated systems and surpasses the computational abilities and performance of existing academic thermal simulators. For example, the simulation of a processor in a package attached to a heat sink, modeled by a computational grid consisting of over 3 million nodes, takes less than 3 minutes. The MTA is fully parallel and publicly available. *

Adaptive Transient Leakage-Aware Linearised Model for Thermal Analysis of 3-D ICs

“…The proposed methodology is implemented and optimised for speed within the MTA package [4] and evaluated on two circuit floorplans derived from the benchmark suite [21]. All experiments are run on a PC with an Intel i7 4790 processor and 32 GB DRAM.…”

“…These are physical models and electro-thermal duality models. The tools in the former category rely on a numerical solution of the heat equation [2], [3], [4], while the simulators in the latter category rely on the duality between electrical circuits and thermal laws [5], [6]. In general, the duality-based models are faster, but less accurate, and have difficulties to capture heterogeneous features present in modern 3-D ICs.…”

“…by the finite element method (FEM) [8]. In order to improve the computational speed and maintain accuracy, spatio-temporal adaptation is typically considered [3], [4], which assumes both periodic refinement/coarsening of a spatial computational grid and adaptive selection of time steps that closely follow the physics of the problem. Such an approach is implemented in the physical thermal simulator MTA [4], which features both linear and non-linear thermal model, and supports both floorplan and cell level detail.…”

3D-ICE 3.0: Efficient Nonlinear MPSoC Thermal Simulation With Pluggable Heat Sink Models

“…They allow to effectively and economically evaluate different dissipation solutions, as well as to perform design space explorations and to aid the design of thermal control policies. However, existing state-of-the-art thermal simulators [12]- [14] are bound to a limited number of simple dissipation solutions, often just a sink with constant fan speed. These current simulators can be configured to change the chip and sink sizes, materials and heat transfer coefficient towards the ambient, however it is not possible for example to represent an arbitrary sink shape or the effect of variable fan speed, let alone addressing more complicated solutions, such as liquid cooling circuits.…”

Thermal management of 3-D integrated circuits with special structures

“…A machine learning (ML)-based control method was proposed and applied to 3-D ICs with the tier-specific microfluidic heat sink in [19]. In [20], the Manchester Thermal Analyzer (MTA) was used to give fast and highly accurate thermal simulation for the complex physical structure. Alqahtani, et al proposed a simulation flow to accurately simulate TTSV effects on 3D ICs, and found that the peak temperature of 3D Nehalem could be reduced by 5-25% [21].…”