|
|
Comparison of Distribtued File System Features
Tomec, Martin ; Kašpárek, Tomáš (referee) ; Szőke, Igor (advisor)
This work deals with comparsion of distributed filesystems Ceph, Lustre and Kosmos FS. General properties of these systems are summarized here, including method of their installing, configuring and launching. Testing tools and test sets are described in the next part. Performance and abilities of these systems were tested on a network of nine computers. Last part contains analyses and evaluation of results, figuring out bottleneck of these systems. Lustre seems to be the best of these systems.
|
|
|
Optimization of the Distributed I/O Subsystem of the k-Wave Project
Vysocký, Ondřej ; Klepárník, Petr (referee) ; Jaroš, Jiří (advisor)
This thesis deals with an effective solution of the parallel I/O of the k-Wave tool, which is designed for time domain acoustic and ultrasound simulations. k-Wave is a supercomputer application, it runs on a Lustre file system and it requires to be implemented with MPI and stores the data in suitable data format (HDF5). I designed three methods of optimization which fits k-Wave's needs. It uses accumulation and redistribution techniques. In comparison with the native write, every optimization method led to better write speed, up to 13.6GB/s. It is possible to use these methods to optimize every data distributed application with the write speed issue.
|
|
|
Comparison of Parallel File Systems
Pazdera, Michal ; Michal, Bohumil (referee) ; Kašpárek, Tomáš (advisor)
The goal of this thesis was to explore several parallel file systems, and to evaluate their performance under various conditions. The main focus of this assessment were read and write speeds in different workloads, the reliability of each system, and also their ability to protect from data loss. Initially, this thesis introduces eight of the most commonly used parallel file systems. From these, three were selected for further testing: Lustre, GlusterFS, and CephFS. To be able to evaluate their performance accurately, a suite of automated tests was developed. These benchmarks were run for each individual file system in our testing lab. The final part of this work evaluates the results and discusses the features of each of the file systems and their suitability for particular workloads.
|
|
| |
|
|
Optimization of the Distributed I/O Subsystem of the k-Wave Project
Vysocký, Ondřej ; Klepárník, Petr (referee) ; Jaroš, Jiří (advisor)
This thesis deals with an effective solution of the parallel I/O of the k-Wave tool, which is designed for time domain acoustic and ultrasound simulations. k-Wave is a supercomputer application, it runs on a Lustre file system and it requires to be implemented with MPI and stores the data in suitable data format (HDF5). I designed three methods of optimization which fits k-Wave's needs. It uses accumulation and redistribution techniques. In comparison with the native write, every optimization method led to better write speed, up to 13.6GB/s. It is possible to use these methods to optimize every data distributed application with the write speed issue.
|
|
|
Optimization of the Distributed I/O Subsystem of the k-Wave Project
Vysocký, Ondřej ; Hrbáček, Radek (referee) ; Jaroš, Jiří (advisor)
This thesis deals with an effective solution of parallel writing of variable amounts of data on the Lustre file system. The work will be used by the k-Wave project designed for time domain acoustic and ultrasound simulations. Since the simulation is computationally and data intensive, the project requires to be implemented with libraries for parallel computig (Open MPI) and large data processing (HDF5) and it must run on a supercomputer. The application is implemented in C and uses previously mentioned libraries. The proper settings of the Lustre file system leads to the peak write bandwith of 2.5 GB/s that corresponds to a speedup factor of 5 compared to the reference settings. The data aggregation improved the write bandwidth by a factor of 3 compared to a naive version. Here, the achieved I/O bandwidth for certain block sizes hits the limits of the Anselm I/O subsytem (3GB/s).
|
|
|
Comparison of Parallel File Systems
Pazdera, Michal ; Michal, Bohumil (referee) ; Kašpárek, Tomáš (advisor)
The goal of this thesis was to explore several parallel file systems, and to evaluate their performance under various conditions. The main focus of this assessment were read and write speeds in different workloads, the reliability of each system, and also their ability to protect from data loss. Initially, this thesis introduces eight of the most commonly used parallel file systems. From these, three were selected for further testing: Lustre, GlusterFS, and CephFS. To be able to evaluate their performance accurately, a suite of automated tests was developed. These benchmarks were run for each individual file system in our testing lab. The final part of this work evaluates the results and discusses the features of each of the file systems and their suitability for particular workloads.
|
|
|
Comparison of Distribtued File System Features
Tomec, Martin ; Kašpárek, Tomáš (referee) ; Szőke, Igor (advisor)
This work deals with comparsion of distributed filesystems Ceph, Lustre and Kosmos FS. General properties of these systems are summarized here, including method of their installing, configuring and launching. Testing tools and test sets are described in the next part. Performance and abilities of these systems were tested on a network of nine computers. Last part contains analyses and evaluation of results, figuring out bottleneck of these systems. Lustre seems to be the best of these systems.
|
|
| |
guest ::
