Related papers: Disaggregated Memory with SmartNIC Offloading: a C…

The success of DNNs and their high computational requirements pushed for large codesign efforts aiming at DNN acceleration. Since DNNs can be represented as static computational graphs, static memory allocation and tiling are two crucial…

Memory is a critical design consideration in current data-intensive DNN accelerators, as it profoundly determines energy consumption, bandwidth requirements, and area costs. As DNN structures become more complex, a larger on-chip memory…

Deploying Deep Neural Networks (DNNs) on different hardware platforms is challenging due to varying resource constraints. Besides handcrafted approaches aiming at making deep models hardware-friendly, Neural Architectures Search is rising…

The ever-increasing computation complexity of fast-growing Deep Neural Networks (DNNs) has requested new computing paradigms to overcome the memory wall in conventional Von Neumann computing architectures. The emerging Computing-In-Memory…

Graph Neural Networks (GNNs) have demonstrated a great potential in a variety of graph-based applications, such as recommender systems, drug discovery, and object recognition. Nevertheless, resource-efficient GNN learning is a rarely…

Edge intelligence has arisen as a promising computing paradigm for supporting miscellaneous smart applications that rely on machine learning techniques. While the community has extensively investigated multi-tier edge deployment for…

This extended report presents DDS, a novel disaggregated storage architecture enabled by emerging networking hardware, namely DPUs (Data Processing Units). DPUs can optimize the latency and CPU consumption of disaggregated storage servers.…

Multi-pod systolic arrays are emerging as the architecture of choice in DNN inference accelerators. Despite their potential, designing multi-pod systolic arrays to maximize effective throughput/Watt (i.e., throughput/Watt adjusted when…

The in-memory graph layout or organization has a considerable impact on the time and energy efficiency of distributed memory graph computations. It affects memory locality, inter-task load balance, communication time, and overall memory…

This paper examines disaggregated data center architectures from the perspective of the applications that would run on these data centers, and challenges the abstractions that have been proposed to date. In particular, we argue that…

Multi-scale deformable attention (MSDA) is a flexible and powerful feature extraction mechanism for visual tasks, but its random-access grid sampling strategy poses significant optimization challenges, especially on domain-specific…

The demand for executing Deep Neural Networks (DNNs) with low latency and minimal power consumption at the edge has led to the development of advanced heterogeneous Systems-on-Chips (SoCs) that incorporate multiple specialized computing…

Convolutional Neural Networks (CNNs) have demonstrated their effectiveness in numerous vision tasks. However, their high processing requirements necessitate efficient hardware acceleration to meet the application's performance targets. In…

Recent advances in integrated photonics enable the implementation of reconfigurable, high-bandwidth, and low energy-per-bit interconnects in next-generation data centers. We propose and evaluate an Optically Connected Memory (OCM)…

Non-Orthogonal Multiple Access (NOMA) is at the heart of a paradigm shift towards non-orthogonal communication due to its potential to scale well in massive deployments. Nevertheless, the overhead of channel estimation remains a key…

Neuro-symbolic AI is gaining traction in domains such as large language models, scientific discovery, and autonomous systems due to its ability to combine perception with structured reasoning. However, its deployment is often constrained by…

Deep learning has shown impressive performance in semantic segmentation, but it is still unaffordable for resource-constrained mobile devices. While offloading computation tasks is promising, the high traffic demands overwhelm the limited…

To support massive connectivity and boost spectral efficiency for internet of things (IoT), a downlink scheme combining virtual multiple-input multiple-output (MIMO) and nonorthogonal multiple access (NOMA) is proposed. All the…

Many distributed applications implement complex data flows and need a flexible mechanism for routing data between producers and consumers. Recent advances in programmable network interface cards, or SmartNICs, represent an opportunity to…

Graphics processing units (GPUs) can improve deep neural network inference throughput via batch processing, where multiple tasks are concurrently processed. We focus on novel scenarios that the energy-constrained mobile devices offload…