Mochi: Composing data services for high-performance computing environments
Technology enhancements and the growing breadth of application workflows running on
high-performance computing (HPC) platforms drive the development of new data services …
high-performance computing (HPC) platforms drive the development of new data services …
Reimagining codesign for advanced scientific computing: Report for the ascr workshop on reimagining codesign
In March 2021, the US Department of Energy's Advanced Scientific Computing Research
program convened the Workshop on Reimagining Codesign. The workshop, also known as …
program convened the Workshop on Reimagining Codesign. The workshop, also known as …
Accelerating hdf5 i/o for exascale using daos
J Soumagne, J Henderson, M Chaarawi… - … on Parallel and …, 2021 - ieeexplore.ieee.org
The Hierarchical Data Format 5 (HDF5) has long been defined as one of the most prominent
data models, binary file formats and I/O libraries for storing and managing scientific data …
data models, binary file formats and I/O libraries for storing and managing scientific data …
Methodology for the rapid development of scalable HPC data services
Growing evidence in the scientific computing community indicates that parallel file systems
are not sufficient for all HPC storage workloads. This realization has motivated extensive …
are not sufficient for all HPC storage workloads. This realization has motivated extensive …
Extending composable data services into smartnics
C Ulmer, J Liu, C Maltzahn… - 2023 IEEE International …, 2023 - ieeexplore.ieee.org
Advanced scientific-computing workflows rely on composable data services to migrate data
between simulation and analysis jobs that run in parallel on high-performance computing …
between simulation and analysis jobs that run in parallel on high-performance computing …
Using a robust metadata management system to accelerate scientific discovery at extreme scales
M Lawson, J Lofstead - … Workshop on Parallel Data Storage & …, 2018 - ieeexplore.ieee.org
Our previous work, which can be referred to as EMPRESS 1.0, showed that rich metadata
management provides a relatively low-overhead approach to facilitating insight from scale …
management provides a relatively low-overhead approach to facilitating insight from scale …
EMPRESS: Accelerating Scientific Discovery through Descriptive Metadata Management
M Lawson, W Gropp, J Lofstead - ACM Transactions on Storage, 2022 - dl.acm.org
High-performance computing scientists are producing unprecedented volumes of data that
take a long time to load for analysis. However, many analyses only require loading in the …
take a long time to load for analysis. However, many analyses only require loading in the …
RaDD runtimes: Radical and different distributed runtimes with smartnics
As network speeds increase, the overhead of processing incoming messages is becoming
onerous enough that many manufacturers now provide network interface cards (NICs) with …
onerous enough that many manufacturers now provide network interface cards (NICs) with …
Opportunistic Query Execution on SmartNICs for Analyzing In-Transit Data
J Liu, C Maltzahn, C Ulmer - 2023 IEEE High Performance …, 2023 - ieeexplore.ieee.org
High-performance computing (HPC) systems researchers have proposed using current,
programmable network interface cards (or SmartNICs) to offload data management services …
programmable network interface cards (or SmartNICs) to offload data management services …
Benesh: a Framework for Choreographic Coordination of In Situ Workflows
The growing scale of high-performance computing systems increasingly enables scientists
to develop more complex applications as in situ workflows composed of coupled simulation …
to develop more complex applications as in situ workflows composed of coupled simulation …