Accelerating HDF5 I/O for Exascale Using DAOS

Soumagne, Jérôme; Henderson, Jordan; Chaarawi, Mohamad; Fortner, Neil; Breitenfeld, Scot; Lu, Songyu; Robinson, Dana; Pourmal, Elena; Lombardi, Johann

doi:10.1109/tpds.2021.3097884

Cited by 11 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A recent study [20] explored the performance of HDF5 over the DAOS object interface. This object-centric design enabled HDF5 to transition away from traditional blockbased storage, thereby circumventing the constraints posed by POSIX.…”

Section: H Warpx Laser Wakefield Simulationmentioning

confidence: 99%

Optimizing Metadata Exchange: Leveraging DAOS for ADIOS Metadata I/O

Venkatesh,

Eisenhauer,

Podhorszki

et al. 2024

ISC High Performance 2024 Research Paper Proceedings (39th International Conference)

View full text Add to dashboard Cite

In HPC I/O middleware like the Adaptable I/O System (ADIOS) often mediates data transfers between applications. The metadata I/O generated by such systems often presents significant scaling and performance limitations. This work seeks improvement opportunities for metadata I/O by leveraging the DAOS storage systems, a recent storage system solution deployed on high-end systems such as the Aurora supercomputer. We investigate the tradeoffs and the design space for integrating I/O engines for the ADIOS middleware based on the different storage mechanisms supported by DAOS. We present a new DAOS-Array-ChunkSize-aligned engine which provides up to 2.3× improved performance than when using the existing DAOS-POSIX interface, without requiring any application modifications.

show abstract

Section: H Warpx Laser Wakefield Simulationmentioning

confidence: 99%

Optimizing Metadata Exchange: Leveraging DAOS for ADIOS Metadata I/O

Venkatesh,

Eisenhauer,

Podhorszki

et al. 2024

ISC High Performance 2024 Research Paper Proceedings (39th International Conference)

View full text Add to dashboard Cite

show abstract

“…Examples include images of varying sizes (one megabyte to several hundred megabytes), matrices based on integers, floats or double precision numbers, or very long text files consisting of DNA sequences. Previously, efforts like ROOT (Antcheva et al, 2011), HDF5 (Soumagne et al, 2022) and similar file formats have been used to exploit the underlying structure to improve the overall application's performance. These successful efforts clearly highlight that scientific datasets need special attention and the performance of the scientific applications can be significantly improved by exploiting the underlying structure in the datasets.…”

Section: Scientific Datasetsmentioning

confidence: 99%

Data Management of Scientific Applications in a Reinforcement Learning-Based Hierarchical Storage System

Zhang

Gupta

Rodríguez

et al. 2023

Preprint

View full text Add to dashboard Cite

“…Previous results for local and distributed persistent memory file systems with NVM are also promising for several real-world applications, exemplifying significant improvements over current state-of-the-art NVMe SSD storage. However, most of the benchmarks employed [21], [22], [23], [24], [25] are not scientific workflows or are not based on common stressors for scientific patterns in supercomputing such as BTIO [26] with all of its variants under MPI. Furthermore, these works do not compare local and distributed storage over RDMA, and they also do not compare to common storage devices such as HDD or SATA-SSD, which are still very common in largescale distributed storage systems.…”

Section: Related Workmentioning

confidence: 99%

Assessing the Use Cases of Persistent Memory in High-Performance Scientific Computing

Yehonatan¹,

Snir²,

Rusanovsky³

et al. 2021

Preprint

View full text Add to dashboard Cite

As the High Performance Computing (HPC) world moves towards the Exa-Scale era, huge amounts of data should be analyzed, manipulated and stored. In the traditional storage/memory hierarchy, each compute node retains its data objects in its local volatile DRAM. Whenever the DRAM's capacity becomes insufficient for storing this data, the computation should either be distributed between several compute nodes, or some portion of these data objects must be stored in a non-volatile block device such as a hard disk drive (HDD) or an SSD storage device. These standard block devices offer large and relatively cheap non-volatile storage, but their access times are orders-ofmagnitude slower than those of DRAM. Optane™ DataCenter Persistent Memory Module (DCPMM) [1], a new technology introduced by Intel, provides non-volatile memory that can be plugged into standard memory bus slots (DDR DIMMs) and therefore be accessed much faster than standard storage devices. In this work, we present and analyze the results of a comprehensive performance assessment of several ways in which DCPMM can 1) replace standard storage devices, and 2) replace or augment DRAM for improving the performance of HPC scientific computations. To achieve this goal, we have configured an HPC system such that DCPMM can service I/O operations of scientific applications, replace standard storage devices and file systems (specifically for diagnostics and checkpoint-restarting), and serve for expanding applications' main memory. We focus on keeping the scientific codes with as few changes as possible, while allowing them to access the NVM transparently as if they access persistent storage. Our results show that DCPMM allows scientific applications to fully utilize nodes' locality by providing them with sufficiently-large main memory. Moreover, it can also be used for providing a high-performance replacement for persistent storage. Thus, the usage of DCPMM has the potential of replacing standard HDD and SSD storage devices in HPC architectures and enabling a more efficient platform for modern supercomputing applications. The source code used by this work, as well as the benchmarks and other relevant sources, are available at: https://github.com/ Scientific-Computing-Lab-NRCN/StoringStorage.

show abstract

Accelerating HDF5 I/O for Exascale Using DAOS

Cited by 11 publications

References 15 publications

Optimizing Metadata Exchange: Leveraging DAOS for ADIOS Metadata I/O

Optimizing Metadata Exchange: Leveraging DAOS for ADIOS Metadata I/O

Data Management of Scientific Applications in a Reinforcement Learning-Based Hierarchical Storage System

Assessing the Use Cases of Persistent Memory in High-Performance Scientific Computing

Contact Info

Product

Resources

About