Predicting performance of non-contiguous i/o with machine learning
Data sieving in ROMIO promises to optimize individual non-contiguous I/O. However,
making the right choice and parameterizing its buffer size accordingly are non-trivial tasks …
making the right choice and parameterizing its buffer size accordingly are non-trivial tasks …
Simulating parallel programs on application and system level
JM Kunkel - Computer Science-Research and Development, 2013 - Springer
Understanding the measured performance of parallel applications in real systems is difficult—
with the aim to utilize the resources available, optimizations deployed in hardware and …
with the aim to utilize the resources available, optimizations deployed in hardware and …
Using simulation to validate performance of MPI (-IO) implementations
JM Kunkel - … Conference, ISC 2013, Leipzig, Germany, June 16-20 …, 2013 - Springer
Parallel file systems and MPI implementations aim to exploit available hardware resources
in order to achieve optimal performance. Since performance is influenced by many …
in order to achieve optimal performance. Since performance is influenced by many …
Optimizations for two-phase collective I/O
The performance of parallel distributed file systems suffers from many clients executing a
large number of operations in parallel, because the I/O subsystem can be easily …
large number of operations in parallel, because the I/O subsystem can be easily …
An implementation of parallel file distribution in an agent hierarchy
M Fukuda, J Miyauchi - The Journal of Supercomputing, 2009 - Springer
PC grid is a cost-effective grid-computing platform that attracts users by allocating to their
massively parallel applications as many desktop computers as requested. However, a …
massively parallel applications as many desktop computers as requested. However, a …
Simulation-aided performance evaluation of server-side input/output optimizations
The performance of parallel distributed file systems suffers from many clients executing a
large number of operations in parallel, because the I/O subsystem can be easily …
large number of operations in parallel, because the I/O subsystem can be easily …
A task-pool parallel i/o paradigm for an i/o intensive application
J Li, L Yan, Z Gao, D Hei - 2009 IEEE International Symposium …, 2009 - ieeexplore.ieee.org
In regards to applications like 3D seismic migration, it is quite important to improve the I/O
performance within an cluster computing system. Such seismic data processing applications …
performance within an cluster computing system. Such seismic data processing applications …
Simulation-Aided Performance Evaluation of Input/Output Optimizations for Distributed Systems
M Kuhn - 2009 - archiv.ub.uni-heidelberg.de
The performance of parallel cluster file systems suffers from many clients executing a large
number of operations in parallel, because the I/O subsystem can be easily overwhelmed by …
number of operations in parallel, because the I/O subsystem can be easily overwhelmed by …
Design and Implementation of Automatic Astronomical Spectra Data Processing System
FL Meng, JC Pan - 2015 International Industrial Informatics and …, 2015 - atlantis-press.com
Astronomical observation produces a large amount of data. The effective storage of the data
is an important work. In this paper, the automatic processing of this kind of data was studied …
is an important work. In this paper, the automatic processing of this kind of data was studied …
[PDF][PDF] New scalable storage techniques for high-performance computing (TEALES) TIN2007-63092
J Carretero - ia.urjc.es
Nowadays there is a great interest on developing new technologies for high-performance
computing using grid and cluster environment integrated together. Storage systems are …
computing using grid and cluster environment integrated together. Storage systems are …