MPI windows on storage for HPC applications

Rivas-Gomez, Sergio; Gioiosa, Roberto; Kestor, Gokcen; Narasimhamurthy, Sai; Laure, Erwin

doi:10.1145/3127024.3127034

Cited by 7 publications

(7 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Social Knowledge Management is production management and dissemination of knowledge, research, and socially promoted epistemological models of the class [13]. Knowledge development in society involves the participation of social actors themselves and Trans Discipline (which is not equivalent to interdisciplinarity).…”

Section: Social Knowledge Managementmentioning

confidence: 99%

Factors Influencing Social Knowledge Management in Social Society: A Systematic Literature Review

Fernando

Meyliana

Hidayanto³

et al. 2020

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

Knowledge is important now for the development of social society; it is necessary for knowledge management. Knowledge management (KM) aims to support the creation, transfer, and application of knowledge in social societies. This fact illustrates that in the management of social knowledge, the role of social communities is very important and is influenced by factors in the process. With this, this study will look for theories and factors that influence KM social interactions that occur in social societies. The method used is a systematic literature review. The results of this study found theories and factors that influence KM in social societies.

show abstract

Section: Social Knowledge Managementmentioning

confidence: 99%

Factors Influencing Social Knowledge Management in Social Society: A Systematic Literature Review

Fernando

Meyliana

Hidayanto³

et al. 2020

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

show abstract

“…Moreover, our design can support a variety of back stores. For instance, remote memory paging that fetches data from a memory server or compute node [6,14] could be easily integrated into UMap by providing a new store object.…”

Section: Related Workmentioning

confidence: 99%

UMap: Enabling Application-driven Optimizations for Page Management

McFadden

Green

Iwabuchi

et al. 2019

2019 IEEE/ACM Workshop on Memory Centric High Performance Computing (MCHPC)

View full text Add to dashboard Cite

Leadership supercomputers feature a diversity of storage, from node-local persistent memory and NVMe SSDs to network-interconnected ash memory and HDD. Memory mapping les on dierent tiers of storage provides a uniform interface in applications. However, system-wide services like mmap are optimized for generality and lack exibility for enabling application-specic optimizations. In this work, we present UMap to enable user-space page management that can be easily adapted to access patterns in applications and storage characteristics. UMap uses the userfaultfd mechanism to handle page faults in multi-threaded applications eciently. By providing a data object abstraction layer, UMap is extensible to support various backing stores. The design of UMap supports dynamic load balancing and I/O decoupling for scalable performance. UMap also uses application hints to improve the selection of caching, prefetching, and eviction policies. We evaluate UMap in ve benchmarks and real applications on two systems. Our results show that leveraging application knowledge for page management could substantially improve performance. On average, UMap achieved 1.25 to 2.5 times improvement using the adapted congurations compared to the system service.

show abstract

“…Details about the semantics of operations on MPI storage windows and the implementation are provided in Ref. [23]. can be attached to a data stream.…”

Section: Higher-level I/o Interfacesmentioning

confidence: 99%

The SAGE project: a storage centric approach for exascale computing

Narasimhamurthy

Danilov

et al. 2018

Proceedings of the 15th ACM International Conference on Computing Frontiers

Self Cite

View full text Add to dashboard Cite

SAGE (Percipient StorAGe for Exascale Data Centric Computing) is a European Commission funded project towards the era of Exascale computing. Its goal is to design and implement a Big Data/Extreme Computing (BDEC) capable infrastructure with associated software stack. The SAGE system follows a storage centric approach as it is capable of storing and processing large data volumes at the Exascale regime.SAGE addresses the convergence of Big Data Analysis and HPC in an era of next-generation data centric computing. This convergence is driven by the proliferation of massive data sources, such as large, dispersed scientific instruments and sensors where data needs to be processed, analyzed and integrated into simulations to derive scientific and innovative insights. A first prototype of the SAGE system has been been implemented and installed at the Jülich Supercomputing Center. The SAGE storage system consists of multiple types of storage device technologies in a multi-tier I/O hierarchy, including flash, disk, and non-volatile memory technologies.The main SAGE software component is the Seagate Mero Object Storage that is accessible via the Clovis API and higher level interfaces.The SAGE project also includes scientific applications for the validation of the SAGE concepts.The objective of this paper is to present the SAGE project concepts, the prototype of the SAGE platform and discuss the software architecture of the SAGE system.

show abstract

MPI windows on storage for HPC applications

Cited by 7 publications

References 34 publications

Factors Influencing Social Knowledge Management in Social Society: A Systematic Literature Review

Factors Influencing Social Knowledge Management in Social Society: A Systematic Literature Review

UMap: Enabling Application-driven Optimizations for Page Management

The SAGE project: a storage centric approach for exascale computing

Contact Info

Product

Resources

About