Related papers: Input-Based Ensemble-Learning Method for Dynamic M…

Serverless Function-as-a-Service (FaaS) platforms provide applications with resources that are highly elastic, quick to instantiate, accounted at fine granularity, and without the need for explicit runtime resource orchestration. This…

Serverless technologies, also known as FaaS (Function as a Service), are promoted as solutions that provide dynamic scalability, speed of development, cost-per-consumption model, and the ability to focus on the code while taking attention…

We focus on the problem of Personalized Federated Continual Learning (PFCL): a group of distributed clients, each with a sequence of local tasks on arbitrary data distributions, collaborate through a central server to train a personalized…

High performance is needed in many computing systems, from batch-managed supercomputers to general-purpose cloud platforms. However, scientific clusters lack elastic parallelism, while clouds cannot offer competitive costs for…

Serverless computing abstracts away server management, enabling automatic scaling, efficient resource utilization, and cost-effective pricing models. However, despite these advantages, it faces the significant challenge of cold-start…

Serverless computing offers attractive scalability, elasticity and cost-effectiveness. However, constraints on memory, CPU and function runtime have hindered its adoption for data-intensive applications and machine learning (ML) workloads.…

Serverless computing is transforming cloud application development, but the performance-cost trade-offs of control plane designs remain poorly understood due to a lack of open, cross-platform benchmarks and detailed system analyses. In this…

The memory system of a modern embedded processor consumes a large fraction of total system energy. We explore a range of different configuration options and show that a reconfigurable design can make better use of the resources available to…

Fault localization, the process of identifying the software components responsible for failures, is essential but often time-consuming. Recent advances in Large Language Models (LLMs) have enabled fault localization without extensive defect…

Serverless execution and most notably the Function as a Service (FaaS) model got quite some attention during the recent years. As of today, all commercial and open source implementations follow the common practice of keeping the execution…

Federated learning is a promising paradigm that allows multiple clients to collaboratively train a model without sharing the local data. However, the presence of heterogeneous devices in federated learning, such as mobile phones and IoT…

Memory-augmented Large Language Models (LLMs) are essential for developing capable, long-term AI agents. Recently, applying Reinforcement Learning (RL) to optimize memory operations, such as extraction, updating, and retrieval, has emerged…

Integrating GPUs into serverless computing platforms is crucial for improving efficiency. However, existing solutions for GPU-enabled serverless computing platforms face two significant problems due to coarse-grained GPU management: long…

Large Language Models (LLMs) have become an essential infrastructure for Artificial General Intelligence (AGI), yet their lack of well-defined memory management systems hinders the development of long-context reasoning, continual…

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…

Serverless computing is an emerging cloud computing paradigm for developing applications at the function level, known as serverless functions. Due to the highly dynamic execution environment, multiple identical runs of the same serverless…

Current proprietary and open-source serverless platforms follow opinionated, hardcoded scheduling policies to deploy the functions to be executed over the available workers. Such policies may decrease the performance and the security of the…

The exponential growth of data-intensive machine learning workloads has exposed significant limitations in conventional GPU-accelerated systems, especially when processing datasets exceeding GPU DRAM capacity. We propose MQMS, an augmented…

Agentic workflows that use autonomous AI Agents powered by Large Language Models (LLMs) and Model Context Protocol (MCP) servers is rapidly rising. This introduces challenges in scalable cloud deployment and state management. Traditional…

In today's production machine learning (ML) systems, models are continuously trained, improved, and deployed. ML design and training are becoming a continuous workflow of various tasks that have dynamic resource demands. Serverless…