arXiv subjects

Valeria Cardellini

Publications and source records attributed to Valeria Cardellini.

6 recordsLinked to original sources

arXiv2026

A Periodic Space of Distributed Computing: Vision & Framework

Advances in networking and computing technologies throughout the early decades of the 21st century have transformed long-standing dreams of pervasive communication and computation into reality. These technologies now form a rapidly evolving and increasingly complex global infrastructure that will underpin the next aspiration of computing: supporting intelligent systems with human-level or even superhuman capabilities. We examine how today's distributed computing landscape can evolve to meet the demands of future users, intelligent systems, and emerging application domains. We propose a "periodic framework" for characterizing the distributed computing landscape, inspired by the systematic structure and explanatory power of the "periodic table" in chemistry. This framework provides a structured way to describe, compare, and reason about the behaviors and design choices of different distributed computing solutions. Using this framework, we can identify patterns in key system properties, such as responsiveness and availability, across the distributed computing landscape. We also explain how the framework can help in predicting future trajectories in the field. Lastly, we synthesize insights from leading researchers worldwide regarding the desired properties, design principles, and implications of emerging areas in the forthcoming distributed computing landscape and in relation to the periodic framework. Together, these perspectives shed light on the considerations that will shape the distributed computing landscape underpinning future intelligent systems.

cs.DC↗

arXiv2025

A Survey on Deep Learning Hardware Accelerators for Heterogeneous HPC Platforms

Recent trends in deep learning (DL) have made hardware accelerators essential for various high-performance computing (HPC) applications, including image classification, computer vision, and speech recognition. This survey summarizes and classifies the most recent developments in DL accelerators, focusing on their role in meeting the performance demands of HPC applications. We explore cutting-edge approaches to DL acceleration, covering not only GPU- and TPU-based platforms but also specialized hardware such as FPGA- and ASIC-based accelerators, Neural Processing Units, open hardware RISC-V-based accelerators, and co-processors. This survey also describes accelerators leveraging emerging memory technologies and computing paradigms, including 3D-stacked Processor-In-Memory, non-volatile memories like Resistive RAM and Phase Change Memories used for in-memory computing, as well as Neuromorphic Processing Units, and Multi-Chip Module-based accelerators. Furthermore, we provide insights into emerging quantum-based accelerators and photonics. Finally, this survey categorizes the most influential architectures and technologies from recent years, offering readers a comprehensive perspective on the rapidly evolving field of deep learning acceleration.

cs.AR↗

arXiv2024

Efficient Probabilistic Workflow Scheduling for IaaS Clouds

The flexibility and the variety of computing resources offered by the cloud make it particularly attractive for executing user workloads. However, IaaS cloud environments pose non-trivial challenges in the case of workflow scheduling under deadlines and monetary cost constraints. Indeed, given the typical uncertain performance behavior of cloud resources, scheduling algorithms that assume deterministic execution times may fail, thus requiring probabilistic approaches. However, existing probabilistic algorithms are computationally expensive, mainly due to the greater complexity of the workflow scheduling problem in its probabilistic form, and they hardily scale with the size of the problem instance. In this article, we propose EPOSS, a novel workflow scheduling algorithm for IaaS cloud environments based on a probabilistic formulation. Our solution blends together the low execution latency of state-of-the-art scheduling algorithms designed for the case of deterministic execution times and the capability to enforce probabilistic constraints.Designed with computational efficiency in mind, EPOSS achieves one to two orders lower execution times in comparison with existing probabilistic schedulers. Furthermore, it ensures good scaling with respect to workflow size and number of heterogeneous virtual machine types offered by the IaaS cloud environment. We evaluated the benefits of our algorithm via an experimental comparison over a variety of workloads and characteristics of IaaS cloud environments.

cs.DC↗

arXiv2024

Autonomy Loops for Monitoring, Operational Data Analytics, Feedback, and Response in HPC Operations

Many High Performance Computing (HPC) facilities have developed and deployed frameworks in support of continuous monitoring and operational data analytics (MODA) to help improve efficiency and throughput. Because of the complexity and scale of systems and workflows and the need for low-latency response to address dynamic circumstances, automated feedback and response have the potential to be more effective than current human-in-the-loop approaches which are laborious and error prone. Progress has been limited, however, by factors such as the lack of infrastructure and feedback hooks, and successful deployment is often site- and case-specific. In this position paper we report on the outcomes and plans from a recent Dagstuhl Seminar, seeking to carve a path for community progress in the development of autonomous feedback loops for MODA, based on the established formalism of similar (MAPE-K) loops in autonomous computing and self-adaptive systems. By defining and developing such loops for significant cases experienced across HPC sites, we seek to extract commonalities and develop conventions that will facilitate interoperability and interchangeability with system hardware, software, and applications across different sites, and will motivate vendors and others to provide telemetry interfaces and feedback hooks to enable community development and pervasive deployment of MODA autonomy loops.

cs.DC↗

arXiv2023

A Survey on Design Methodologies for Accelerating Deep Learning on Heterogeneous Architectures

Given their increasing size and complexity, the need for efficient execution of deep neural networks has become increasingly pressing in the design of heterogeneous High-Performance Computing (HPC) and edge platforms, leading to a wide variety of proposals for specialized deep learning architectures and hardware accelerators. The design of such architectures and accelerators requires a multidisciplinary approach combining expertise from several areas, from machine learning to computer architecture, low-level hardware design, and approximate computing. Several methodologies and tools have been proposed to improve the process of designing accelerators for deep learning, aimed at maximizing parallelism and minimizing data movement to achieve high performance and energy efficiency. This paper critically reviews influential tools and design methodologies for Deep Learning accelerators, offering a wide perspective in this rapidly evolving field. This work complements surveys on architectures and accelerators by covering hardware-software co-design, automated synthesis, domain-specific compilers, design space exploration, modeling, and simulation, providing insights into technical challenges and open research directions.

cs.AR↗

arXiv2022

An Intrusion Response System utilizing Deep Q-Networks and System Partitions

Intrusion Response is a relatively new field of research. Recent approaches for the creation of Intrusion Response Systems (IRSs) use Reinforcement Learning (RL) as a primary technique for the optimal or near-optimal selection of the proper countermeasure to take in order to stop or mitigate an ongoing attack. However, most of them do not consider the fact that systems can change over time or, in other words, that systems exhibit a non-stationary behavior. Furthermore, stateful approaches, such as those based on RL, suffer the curse of dimensionality, due to a state space growing exponentially with the size of the protected system. In this paper, we introduce and develop an IRS software prototype, named irs-partition. It leverages the partitioning of the protected system and Deep Q-Networks to address the curse of dimensionality by supporting a multi-agent formulation. Furthermore, it exploits transfer learning to follow the evolution of non-stationary systems.

cs.CR↗