Related papers: Efficient Data-Plane Memory Scheduling for In-Netw…

5G networks are expected to be highly energy efficient, with a 10 times lower consumption than today's systems. An effective way to achieve such a goal is to act on the backhaul network by controlling the nodes operational state and the…

IoT applications increasingly rely on on-device AI accelerators to ensure high performance, especially in low-connectivity and safety-critical scenarios. However, the limited on-chip memory of these accelerators forces inference runtimes to…

Deep Learning (DL) training platforms are built by interconnecting multiple DL accelerators (e.g., GPU/TPU) via fast, customized interconnects with 100s of gigabytes (GBs) of bandwidth. However, as we identify in this work, driving this…

To meet the demands of wireless sensor networks (WSNs) where data are usually aggregated at a single source prior to transmitting to any distant user, there is a need to establish a tree structure inside to aggregate data. In this paper, an…

U-Net models with encoder, decoder, and skip-connections components have demonstrated effectiveness in a variety of vision tasks. The skip-connections transmit fine-grained information from the encoder to the decoder. It is necessary to…

The emergence of Mixture-of-Experts (MoE) has transformed the scaling of large language models by enabling vast model capacity through sparse activation. Yet, converting these performance gains into practical edge deployment remains…

In neural network topologies, algorithms are running on batches of data tensors. The batches of data are typically scheduled onto the computing cores which execute in parallel. For the algorithms running on batches of data, an optimal batch…

Simulated annealing (SA) is a well-known algorithm for solving combinatorial optimization problems. However, the computation time of SA increases rapidly, as the size of the problem grows. Recently, a stochastic simulated annealing (SSA)…

Data loading can dominate deep neural network training time on large-scale systems. We present a comprehensive study on accelerating data loading performance in large-scale distributed training. We first identify performance and scalability…

Mobile-edge computing (MEC) has emerged as a prominent technique to provide mobile services with high computation requirement, by migrating the computation-intensive tasks from the mobile devices to the nearby MEC servers. To reduce the…

Memory-aware network scheduling is becoming increasingly important for deep neural network (DNN) inference on resource-constrained devices. However, due to the complex cell-level and network-level topologies, memory-aware scheduling becomes…

This paper summarizes the ideas and key concepts in MISE (Memory Interference-induced Slowdown Estimation), which was published in HPCA 2013 [97], and examines the work's significance and future potential. Applications running concurrently…

Emergency communications networks require in-network intelligence for timely traffic handling under dynamic demands and runtime constraints. In these environments, packets may need different inference behaviors, and conventional model…

As a consequence of the high variability of load demand and renewable generation, long-term and high-resolution inputs are required for power system expansion planning, making the problem intractable in real-world applications. Time series…

Transformers have become central to natural language processing and large language models, but their deployment at scale faces three major challenges. First, the attention mechanism requires massive matrix multiplications and frequent…

Scheduling the power exchange between a population of heterogeneous distributed energy resources and the corresponding upper-level system is an important control problem in power systems. A key challenge is the large number of (partially…

Practical sequence classification tasks in natural language processing often suffer from low training data availability for target classes. Recent works towards mitigating this problem have focused on transfer learning using embeddings…

Distributed Quantum Computers (DQCs) enable large system sizes by connecting smaller chips via photonic interconnects. DQCs use teleportation to relocate qubits and execute CNOTs between qubits on different chips. However, non-local CNOTs…

Consider the problem of joint uplink scheduling and power allocation. Being inherent to almost any wireless system, this resource allocation problem has received extensive attention. Yet, most common techniques either adopt classical power…

Edge computing has become a very popular service that enables mobile devices to run complex tasks with the help of network-based computing resources. However, edge clouds are often resource-constrained, which makes resource allocation a…