Thorlindur Thorolfsson

Design and CAD for 3D integrated circuits

High density through silicon vias (TSV) can be used to build 3DICs that enable unique applications in computing, signal processing and memory intensive systems. This paper presents several case studies that are uniquely enhanced through 3D implementation, including a 3D CAM, an FFT processor, and a SAR processor. The CAD flow used to implement for these designs is described. 3DIC requires higher fidelity thermal modeling than 2DIC design. The rationale for this requirement is established and a possible solution is presented.

Design automation for a 3DIC FFT processor for synthetic aperture radar: a case study

This work discusses a 1024-point, memory-on-logic 3DIC FFT processor for synthetic aperture radar (SAR), sent to fabrication in the 180 nm MIT Lincoln Labs 3D FDSOI 1.5 V process along with the design flow required to realize it with off-the-shelf commercial 2D tools. The work shows how the vertical dimension can be exploited for novel memory architecture tradeoffs that are not feasible in 2D, reducing the energy consumed per memory operation in the FFT by 60.3%. In comparison to its 2D counterpart, the SAR FFT processor exhibits a 53.0% decrease in average wire length, a 24.6% increase in maximum operating frequency and a 25.3% decrease in total silicon area.

Junction-Level Thermal Analysis of 3-D Integrated Circuits Using High Definition Power Blurring

The degraded thermal path of 3-D integrated circuits (3DICs) makes thermal analysis at the chip-scale an essential part of the design process. Performing an appropriate thermal analysis on such circuits requires a model with junction-level fidelity; however, the computational burden imposed by such a model is tremendous. In this paper, we present enhancements to two thermal modeling techniques for integrated circuits to make them applicable to 3DICs. First, we present a resistive mesh-based approach that improves on the fidelity of prior approaches by constructing a thermal model of the full structure of 3DICs, including the interconnect. Second, we introduce a method for dividing the thermal response caused by a heat load into a high fidelity “near response” and a lower fidelity “far response” in order to implement Power Blurring high definition (HD), a hierarchical thermal simulation approach based on Power Blurring that incorporates the resistive mesh-based models and allows for junction-level accuracy at the full-chip scale. The Power Blurring HD technique yields approximately three orders of magnitude of improvement in memory usage and up to six orders of magnitude of improvement in runtime for a three-tier synthetic aperture radar circuit, as compared to using a full-chip junction-scale resistive mesh-based model. Finally, measurement results are presented showing that Power Blurring high definition (HD) accurately determines the shape of the thermal profile of the 3DIC surface after a correction factor is added to adjust for a discrepancy in the absolute temperature values.

Logic-on-logic 3D integration and placement

In this paper we describe three 3D standard cell placement algorithms, which are: “3D Placement using Sequential Off-the-Shelf 2D Placement Tools”, “True-3D Analytical Placement with mPL” and “3D Placement using Simultaneous 2D Placements with mPL”. We use these algorithms to place three case studies in a real face-to-face 3D integration process. The three case studies are a 2 point FFT butterfly processing element (PE), an Advanced Encryption Standard encryption block (AES) and a multiple-input and multiple-output wireless decoder (MIMO). The placements are then fully routed and compared to 2D placements in terms of performance and power consumption. Using this methodology we show that using 3D face-to-face integration with microbumps in conjunction with the three placement algorithms we can improve the maximum clock speed of AES module by 15.3% and the PE by 22.6%, while reducing the power of the AES module and the PE by 2.6% and 12.9% respectively.

Junction-level thermal extraction and simulation of 3DICs

In 3DICs heat dissipating devices are stacked directly on top of each other leading to a higher heat density than in a comparable 2D chip. 3D integration also moves the majority of active devices further away from the heatsink. This results in a degraded thermal path which makes it more challenging to remove heat from the active devices. Gradient FireBolt was used to perform an appropriate 3D thermal analysis on a 1024-point, memory-on-logic 3DIC FFT processor for synthetic aperture radar (SAR). The chip was simulated with a spatial resolution of 80 nm, and was modeled to include the effect of each line of interconnect, as well as each via and fill structure exactly as drawn in the layout. Large isolated temperature spikes were found near groups of clock buffers at the edge of the SRAMs on the middle tier. It was found that lowering the simulation resolution and using composite thermal conductivities failed to accurately predict the location of these tentpoles.

A low-area flexible MIMO detector for WiFi/WiMAX standards

MIMO wireless technology is required to increase the data rates for a broad range of applications, including low cost mobile devices. In this paper we present a very low area reconfigurable MIMO detector which achieves a high throughput of 103Mbps and uses 27 Kilo Gates when implemented in a commercial 180nm CMOS process. The low area is achieved by the proposed in-place architecture. This architecture implements the K-best algorithm and reduces area 4-fold compared to the widely used multi-stage architecture, while provides reconfigurability in terms of antenna configuration during real-time operation.

Comparative analysis of two 3D integration implementations of a SAR processor

When designing 3DICs there are five major issues that differ from 2D that must receive special attention: power delivery, thermal density, design for test, clock tree design and floorplanning. Power delivery in 3D must receive special attention as 3D designs have larger supply currents flowing through the package power delivery pins, along with a longer power delivery path than in comparable 2D system. Thermal density is an issue as 3D integrated chips will have more heat density and less capacity to remove heat than a comparable 2D chip. 3D clock tree distribution is much more difficult than in 2D because the most commonly used methodologies and design tools are geared towards 2D designs and process variation between the different tiers makes it harder to keep skew, jitter and power consumption down. Design for test is harder in 3D because 3D vias provide another point of failure and post fabrication repairs such as Focused Ion Beam are more difficult to perform in 3D. Finally, floorplanning is drastically different in 3D than in 2D, and the four aforementioned issues must all be taken into account during 3D floorplanning. In this paper, all five design issues are explored in the context of a high-resolution memory-on-logic Synthetic Aperture Radar (SAR) processor. The SAR processor is chosen specifically as it requires a significant amount of memory bandwidth that is best met with the high I/O bandwidth afforded by a 3D process. The issues are examined in the context of two implementations for two different 3D integration processes. The first implementation was done in MIT Lincoln Laboratorys 3D FDSOI 1.5 V three tier process and is currently in fabrication. The second design is currently in the design stage, and will be fabricated in two tiers of Chartered Semiconductors 130 nm process 3D integrated with two tiers of high bandwidth DRAM using Tezzaron Semiconductors vertical interconnection technology.

A 10.35 mW/GFlop stacked SAR DSP unit using fine-grain partitioned 3D integration

In this paper we present a technique for implementing a fine-grain partitioned three-dimensional SAR DSP system using 3D placement of standard cells where only one of the 3D tiers is clocked to reduce clock power. We show how this technique was used to build the first fine-grain partitioned 3D integrated system to be demonstrated with silicon measurements in the literature, which is an ultra efficient floating-point synthetic aperture radar (SAR) DSP processing unit. The processing unit was fabricated in two tiers of GlobalFoundries, 1.5 V 130nm process that were 3D stacked face-to-face by Tezzaron. After fabrication the test chip was measured to consume 4.14 mW of power while running at 40 MHz operating for an operating efficiency of 10.35 mW/GFlop.

A low power 3D integrated FFT engine using hypercube memory division

In this paper we demonstrate a floating point FFT processor that leverages both 3D integration and a hypercube memory division scheme to reduce the power consumption of a 1024 point FFT down to 4.227 μJ. The hypercube memory division scheme lowers the energy per memory access by 59.2% while only increasing the total area required by 16.8%, while using 3D integration reduces the logic power by 5.2%. For comparison, we analyze the amount of power and wire length reduction that can be expected from 3D integration for normal digital logic circuits.

An energy-efficient 64-QAM MIMO detector for emerging wireless standards

A power/area aware design is mandatory for the MIMO (Multi-Input Multi-Output) detectors used in LTE and WiMAX standards. The 64-QAM modulation used in the MIMO detector requires more detection effort compared to the smaller constellation sizes widely implemented in the literature. In this work we propose a new architecture for the K-best detector, which unlike the popular multi-stage architecture used for K-best detectors, implements just one core. Also, we introduce a slight modification to the K-best algorithm that reduces the number of multiplications by 44%, and reduces the total power consumption by 27%, without any noticeable performance degradation. The overall architecture consumes only 24KGate, which is the smallest area compared to the other implementations in the literature. It also results in an at least 4-fold greater throughput-efficiency (Mbps/KiloGate) compared to the other detectors, while consuming a small power. The decoder implemented in a commercial 130nm process provides a data-rate of 107Mbps, and consumes 54.4mW.