ECHOFS: A Scheduler-Guided Temporary Filesystem to Leverage Node-Local NVMS

Miranda, Alberto; Nou, Ramón; Cortés, Toni

doi:10.1109/cahpc.2018.8645894

Cited by 3 publications

(1 citation statement)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The cost of supporting POSIX consistency is becoming increasingly unacceptable due to two key reasons: (1) the rapid growth in the scale of HPC systems, which directly increases the software overhead of maintaining POSIX consistency; (2) the emergence of new storage devices such as solid storage devices (SSDs), which greatly improves I/O latency and bandwidth and makes software overhead more significant. In recent years, many efforts have been made to develop burst buffer (BB) PFSs [9], [10], [11], [12], [13] (especially user-level systems) with relaxed consistency models, but these models were typically defined ambiguously and informally as by-products of their PFS implementations. This leads to three major issues: (1) Performance: It is challenging for system developers to evaluate and compare the effectiveness of different consistency models; (2) Correctness: It is difficult for programmers to reason about their program or check the correctness of their code;…”

Section: Introductionmentioning

confidence: 99%

Recorder 2.0: Efficient Parallel I/O Tracing and Analysis

Wang

Sun

Snir

et al. 2020

2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)

View full text Add to dashboard Cite

The semantics of HPC storage systems are defined by the consistency models to which they abide. Storage consistency models have been less studied than their counterparts in memory systems, with the exception of the POSIX standard and its strict consistency model. The use of POSIX consistency imposes a performance penalty that becomes more significant as the scale of parallel file systems increases and the access time to storage devices, such as node-local solid storage devices, decreases. While some efforts have been made to adopt relaxed storage consistency models, these models are often defined informally and ambiguously as by-products of a particular implementation. In this work, we establish a connection between memory consistency models and storage consistency models and revisit the key design choices of storage consistency models from a high-level perspective. Further, we propose a formal and unified framework for defining storage consistency models and a layered implementation that can be used to easily evaluate their relative performance for different I/O workloads. Finally, we conduct a comprehensive performance comparison of two relaxed consistency models on a range of commonly-seen parallel I/O workloads, such as checkpoint/restart of scientific applications and random reads of deep learning applications. We demonstrate that for certain I/O scenarios, a weaker consistency model can significantly improve the I/O performance. For instance, in small random reads that typically found in deep learning applications, session consistency achieved an 5x improvement in I/O bandwidth compared to commit consistency, even at small scales.

show abstract