A Systematic Literature Review of I/O Optimization in HPC and Cloud Computing Environments

• Manuel Alexander Garcia Napa UFRGS
• Arthur F. Lorenzon UFRGS

Resumo

Efficient data management strategies are increasingly critical in high-performance computing (HPC) and cloud computing environments, where large volumes of data must be processed without introducing performance bottlenecks that can affect scalability. In this scenario, this paper presents a systematic literature review (SLR) investigating the relationship between I/O operations and HPC and cloud environments. Following the PRISMA framework, we selected and analyzed 88 studies that provide information on the current state of I/O research in HPC and the cloud. A well-defined research protocol guided the selection of relevant studies, limiting the review to works published in the last decade to ensure a contemporary understanding of advances and challenges in I/O operations. The analysis of these studies involved a structured categorization, which identified the benchmarks, file systems, and emerging technologies employed. Our discussion also highlights current trends in optimizing I/O performance, offering an overview of the developments in this area of HPC and cloud computing research.

Palavras-chave: Cloud computing, Protocols, Limiting, File systems, High performance computing, Scalability, Conferences, Computer architecture, Market research, Optimization, I/O, systematic literature review, cloud computing, high-performance computing