Related papers: SwitchAgg:A Further Step Towards In-Network Comput…

With the rapid development in wide area networks and low cost, powerful computational resources, grid computing has gained its popularity. With the advent of grid computing, space limitations of conventional distributed systems can be…

Fog computing extends cloud computing technology to the edge of the infrastructure to let IoT applications access objects' data with reduced latency, location awareness and dynamic computation. By displacing workloads from the central cloud…

Asynchronous methods are fundamental for parallelizing computations in distributed machine learning. They aim to accelerate training by fully utilizing all available resources. However, their greedy approach can lead to inefficiencies using…

Accurately estimating workload runtime is a longstanding goal in computer systems, and plays a key role in efficient resource provisioning, latency minimization, and various other system management tasks. Runtime prediction is particularly…

Agentic Reinforcement Learning (RL) enables Large Language Models (LLMs) to perform autonomous decision-making and long-term planning. Unlike standard LLM post-training, agentic RL workloads are highly heterogeneous, combining…

The interconnection network is a crucial subsystem in High-Performance Computing clusters and Data-centers, guaranteeing high bandwidth and low latency to the applications' communication operations. Unfortunately, congestion situations may…

The recent proliferation of Data Grids and the increasingly common practice of using resources as distributed data stores provide a convenient environment for communities of researchers to share, replicate, and manage access to copies of…

Metascheduling in time-triggered architectures has been crucial in adapting to dynamic and unpredictable environments, ensuring the reliability and efficiency of task execution. However, traditional approaches face significant challenges…

Data aggregation is an efficient approach to handle the congestion introduced by a massive number of machine type devices (MTDs). The aggregators not only collect data but also implement scheduling mechanisms to cope with scarce network…

We propose SwitchTT, a multi-task extension to Trajectory Transformer but enhanced with two striking features: (i) exploiting a sparsely activated model to reduce computation cost in multi-task offline model learning and (ii) adopting a…

The rapid evolution of Artificial Intelligence (AI) and Machine Learning (ML) has significantly heightened computational demands, particularly for inference-serving workloads. While traditional cloud-based deployments offer scalability,…

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…

It has been demonstrated that deep neural networks outperform traditional machine learning. However, deep networks lack generalisability, that is, they will not perform as good as in a new (testing) set drawn from a different distribution…

GPU clusters in multi-tenant settings often suffer from underutilization, making GPU-sharing technologies essential for efficient resource use. Among them, NVIDIA Multi-Instance GPU (MIG) has gained traction for providing hardware-level…

Software-hardware co-design is essential for optimizing in-memory computing (IMC) hardware accelerators for neural networks. However, most existing optimization frameworks target a single workload, leading to highly specialized hardware…

Rapid advancements in cloud based platforms providing access to quantum computing capabilities have opened up several challenges for efficient usage of these highly delicate and costly devices. Although most of the current systems use a…

This paper addresses emerging system-level challenges in heterogeneous retrieval-augmented generation (RAG) serving, where complex multi-stage workflows and diverse request patterns complicate efficient execution. We present HedraRAG, a…

The emerging large-scale and data-hungry algorithms require the computations to be delegated from a central server to several worker nodes. One major challenge in the distributed computations is to tackle delays and failures caused by the…

We present a novel mathematical framework for the specification and analysis of fault-resilient distributed protocols and their implementations, with the following components: 1. Transition systems that allow the specification and analysis…

We consider the problem of stragglers in distributed computing systems. Stragglers, which are compute nodes that unpredictably slow down, often increase the completion times of tasks. One common approach to mitigating stragglers is work…