The performance of Deep-Learning (DL) computing frameworks rely on the performance of
data ingestion and checkpointing. In fact, during the training, a considerable high number of …

The I/O subsystem for the Summit supercomputer, No. 1 on the Top500 list, and its
ecosystem of analysis platforms is composed of two distinct layers, namely the in-system …

Scientific computing workloads at HPC facilities have been shifting from traditional
numerical simulations to AI/ML applications for training and inference while processing and …

Data-intensive platforms such as Hadoop and Spark are routinely used to process massive
amounts of data residing on distributed file systems like HDFS. Increasing memory sizes and …

Buffered I/O via page cache is used for file scanning in many cases as page cache can
provide buffering, data aggregation, I/O alignment and prefetching transparently. However …

Flash-based solid-state drives (SSDs) have been widely adopted in various storage
systems, manifesting better performance than their forerunner HDDs. However, the …

I/O efficiency is essential to productivity in scientific computing, especially as many scientific
domains become more data-intensive. Many characterization tools have been used to …

At present, I/O performance analysis requires different tools to characterize individual
components of the I/O subsystem, and institutional I/O expertise is relied upon to translate …

To meet the exascale I/O requirements for the High-Performance Computing (HPC), a new
I/O subsystem, Burst Buffer, based on solid state drives (SSD), has been developed …

Scientific discoveries are increasingly dependent upon the analysis of large volumes of data
from observations and simulations of complex phenomena. Scientists compose the complex …