Related papers: Fine Grain 3D Integration for Microarchitecture De…
Computing is bottlenecked by data. Large amounts of application data overwhelm storage capability, communication capability, and computation capability of the modern machines we design today. As a result, many key applications' performance,…
Developing a packaging scheme that meets all of the requirements for operation of solid-state qubits in a cryogenic environment can be a formidable challenge. In this article, we discuss work being done in our group as well as in the…
Implementing embedded neural network processing at the edge requires efficient hardware acceleration that couples high computational performance with low power consumption. Driven by the rapid evolution of network architectures and their…
Quality diversity algorithms can be used to efficiently create a diverse set of solutions to inform engineers' intuition. But quality diversity is not efficient in very expensive problems, needing 100.000s of evaluations. Even with the…
More computational resources (i.e., more physical qubits and qubit connections) on a superconducting quantum processor not only improve the performance but also result in more complex chip architecture with lower yield rate. Optimizing both…
3D promises a new dimension in composing systems by aggregating chips. Literally. While the most common uses are still tightly connected with its early forms as a packaging technology, new application domains have been emerging. As the…
Photonic technologies have shown a promising way to build high-speed and high-energy-efficiency neural network accelerators. In previously presented photonic neural networks, architectures are mainly designed for fully-connected layers.…
In this work I present a technique of construction and fast evaluation of a family of cubic polynomials for analytic smoothing and graphical rendering of particles trajectories for flows in a generic geometry. The principal result of the…
The trade-off between coarse- and fine-grained locking is a well understood issue in operating systems. Coarse-grained locking provides lower overhead under low contention, fine-grained locking provides higher scalability under contention,…
Packages capable of supporting large arrays of high-coherence superconducting qubits are vital for the realisation of fault-tolerant quantum computers and the necessary high-throughput metrology required to optimise fabrication and…
The inherent trade-off between ultra-low thermal conductivity and high mechanical rigidity in natural materials limits their utility in advanced applications. Inspired by the unique architecture of layered honeycomb structures, this study…
Producing interactive 3D printed objects currently requires laborious 3D design and post-instrumentation with off-the-shelf electronics. Multi-material 3D printing using conductive PLA presents opportunities to mitigate these challenges. We…
The rapid advancement of three-dimensional integrated circuits (3DICs) has heightened the need for early-phase design space exploration (DSE) to minimize design iterations and unexpected challenges. Emphasizing the pre-register-transfer…
The autoregressive decoding in LLMs is the major inference bottleneck due to the memory-intensive operations and limited hardware bandwidth. 3D-stacked architecture is a promising solution with significantly improved memory bandwidth, which…
A modern graphics processing unit (GPU) is able to perform massively parallel scientific computations at low cost. We extend our implementation of the checkerboard algorithm for the two dimensional Ising model [T. Preis et al., J. Comp.…
Today's systems are overwhelmingly designed to move data to computation. This design choice goes directly against at least three key trends in systems that cause performance, scalability and energy bottlenecks: (1) data access from memory…
To match the blooming demand of generative AI workloads, GPU designers have so far been trying to pack more and more compute and memory into single complex and expensive packages. However, there is growing uncertainty about the scalability…
Dense 3D convolutions provide high accuracy for perception but are too computationally expensive for real-time robotic systems. Existing tri-plane methods rely on 2D image features with interpolation, point-wise queries, and implicit MLPs,…
High Performance Computing is an internet based computing which makes computer infrastructure and services available to the user for research purpose. However, an important issue which needs to be resolved before High Performance Computing…
Thermal energy storage using sensible heating of a solid storage medium is a potential low-cost technology for long-duration energy storage. To effectively get heat in and out of the solid material, channels of heat transfer fluid can be…