Robert J. Ribando

Many-core design from a thermal perspective

Air cooling limits have been a major design challenge in recent years for integrated circuits. Multi-core exacerbates thermal challenges because power scales with the number of cores, but also creates new opportunities for temperature-aware design, because multi-core designs offer more design parameters than single-core designs. This paper investigates the relationship between core size and on-chip hot spot temperature and shows that with the same power density, smaller cores are cooler than larger cores due to a spatial low-pass filtering effect of temperature. This phenomenon suggests that designs exploiting low-pass filtering can dissipate more power within the same cooling budget than contemporary designs.

Natural Convection in a Porous Medium: Effects of Confinement, Variable Permeability, and Thermal Boundary Conditions

Two-dimensional numerical calculations are reported for natural convection of a fluid in a porous, horizontal layer heated from below. Effects of the following parameters are examined: rigid (impermeable) and constant-pressure (permeable) upper boundaries; isothermal and uniform heat flux at the lower boundary; and permeabilities which are constant, or which vary with depth to simulate compaction of a porous medium or property variations of real fluids within the medium. Steady-state results are presented for the heat flux distribution on the upper surface, as well as for flow and temperature fields in the interior.

Accurate, Pre-RTL Temperature-Aware Design Using a Parameterized, Geometric Thermal Model

Preventing silicon chips from negative, even disastrous thermal hazards has become increasingly challenging these days; considering thermal effects early in the design cycle is thus required. To achieve this, an accurate yet fast temperature model together with an early-stage, thermally optimized, design flow are needed. In this paper, we present an improved block-based compact thermal model (HotSpot 4.0) that automatically achieves good accuracy even under extreme conditions. The model has been extensively validated with detailed finite-element thermal simulation tools. We also show that properly modeling package components and applying the right boundary conditions are crucial to making full-chip thermal models like HotSpot accurately resemble what happens in the real world. Ignoring or over-simplifying package components can lead to inaccurate temperature estimations and potential thermal hazards that are costly to fix in later designs stages. Such a full-chip and package thermal model can then be incorporated into a thermally optimized design flow where it acts as an efficient communication medium among computer architects, circuit designers and package designers in early microprocessor design stages, to achieve early and accurate design decisions and also faster design convergence. For example, the temperature-leakage interaction can be readily analyzed within such a design flow to predict potential thermal hazards such as thermal runaway.

Differentiating the roles of IR measurement and simulation for power and temperature-aware design

In temperature-aware design, the presence or absence of a heatsink fundamentally changes the thermal behavior with important design implications. In recent years, chip-level infrared (IR) thermal imaging has been gaining popularity in studying thermal phenomena and thermal management, as well as reverse-engineering chip power consumption. Unfortunately, IR thermal imaging needs a peculiar cooling solution, which removes the heatsink and applies an IR-transparent liquid flow over the exposed bare die to carry away the dissipated heat. Because this cooling solution is drastically different from a normal thermal package, its thermal characteristics need to be closely examined. In this paper, we characterize the differences between two cooling configurations—forced air flow over a copper heatsink (AIR-SINK) and laminar oil flow over bare silicon (OIL-SILICON). For the comparison, we modify the HotSpot thermal model by adding the IR-transparent oil flow and the secondary heat transfer path through the package pins, hence modeling what the IR camera actually sees at runtime. We show that OIL-SILICON and AIR-SINK are significantly different in both transient and steady-state thermal responses. OIL-SILICON has a much slower short-term transient response, which makes dynamic thermal management less efficient. In addition, for OIL-SILICON, the direction of oil flow plays an important role by changing hot spot location, thus impacting hot spot identification and thermal sensor placement. These results imply that the power- and temperature-aware design process cannot just rely on IR measurements. Simulation and IR measurement are both needed and are complementary techniques.

Exploring the thermal impact on manycore processor performance

Performance of processors with many simple parallel cores is limited by the serial part of the workload, requiring an asymmetric core organization with one or more aggressive “primary” cores for better serial performance. A primary core introduces power-hungry microarchitectural structures and usually causes severe local hot spots. This paper explores the thermal impact on manycore processor architecture and evaluates its performance. Preliminary results show that thermal constraints reduce performance as expected, but also make performance almost insensitive to the complexity of the primary core across a diverse degrees of parallelism, which greatly reduces design complexity.

Interaction of scaling trends in processor architecture and cooling

It is predicted that two important trends are likely to accompany traditional CMOS semiconductor technology scaling - chip multiprocessors and 3D integration. With the ever-increasing power consumption and the consequent difficulty in heat removal, it is important to consider the limits and implications of different cooling methods for the upcoming many-core and 3D era. In this paper, we consider both technology scaling and many-core architecture scaling trends in conjunction with conventional air cooling and advanced microchannel cooling for both 2D and 3D microprocessors and identify interesting inflection design points down the road.

A theoretical model of gas transport between arterioles and tissue

A theoretical model of CO2 and O2 diffusion between arterioles and tissue was developed to determine if significant transport could occur in precapillary vessels. There is increasing evidence, both theoretical and experimental, that such exchange does occur. Using a model in which CO2 and O2 were coupled through the Bohr and Haldane effects, we quantified the radial and axial transport. We also examined the roles of axial diffusion in the arteriole wall and tissue and capillary structure on the transport. Capillary arrangements investigated included capillaries independent of the arteriole with the entering capillary PCO2 or PO2 equal to a constant, and capillaries branching off along the length of the arteriole with the entering capillary partial pressure equal to the arteriole partial pressure at the given axial location. We found that for CO2 in arterioles with an inner diameter ranging from 200 to 22 microns, the exiting blood was 6 to 45% of the way to complete equilibrium with the surrounding tissue, respectively. For O2, the range was 8 to 25%, respectively. We also determined that axial diffusion in the arteriole wall and tissue has little effect on the transport and that capillary structure can alter tissue PCO2 by as much as 12 mm Hg in the smallest arteriole, but has little effect on O2 transport.

Temperature-to-power mapping

Accurate power maps are useful for power model validation, process variation characterization, leakage estimation, and power optimization, but are hard to measure directly. Deriving power maps from measured thermal maps is the inverse problem of the power-to-temperature mapping, extensively studied through thermal simulation. Until recently this inverse heat conduction problem has received little attention in the microarchitecture research community. This paper first identifies the source of difficulties for the problem. The inverse mapping is then performed by applying constraints from microarchitecture-level observations. The inherent large sensitivity of the resultant power map is minimized through thermal map-filtering and constrained least-squares optimization. Choices of filter parameters and optimization constraints are investigated and their effects are evaluated. Furthermore, the paper highlights the differences between the grid and block modeling in the inverse mapping which were often ignored by previous schemes. The proposed methods reduce the mapping error by more than 10× compared to unoptimized solutions. To our best knowledge this is the first work to quantitatively evaluate and minimize the noise effect in the temperature to power mapping problem at the microarchitecture level for both grid and block mode, and for the steady and transient case.

General numerical scheme for heat exchanger thermal analysis and design

This article describes two personal computer–based teaching modules designed to instruct engineering students in the more fundamental thermal aspects of the analysis and design of heat exchangers. One module applies to shell‐and‐tube exchangers; the other applies to cross‐flow types. Enhanced numerical algorithms solve discretized forms of the governing differential energy‐conservation equations. Graphical user interfaces display temperature distributions along with traditional design and performance measures.

Teaching module for one‐dimensional, transient conduction

This article describes a PC‐based teaching module designed to instruct engineering students in transient one‐dimensional conduction heat transfer analysis. The module is set up to analyze infinite slabs, infinite cylinders, and spheres, i.e., the same geometries for which the Heisler charts have been used for the last half‐century. The finite‐volume method is used to derive the discretized governing equations. The on‐screen display shows the evolution of the temperature distribution in animated form. Extensive help files introduce students to the fundamentals of solving parabolic, partial differential equations numerically so that they can set up a simple calculation themselves for cases not covered by either this algorithm or the Heisler charts.