Improving object management in HPC workflows

Object management represents a substantial fraction of the total computing time in any distributed application, and it also adds complexity in terms of source code. This project proposes and implements a set of features aimed to improve both the usability and performance of distributed applications

Detalles Bibliográficos
Autor: Rodríguez Guasch, Sergio
Tipo de recurso: tesis de maestría
Fecha de publicación:2019
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/167809
Acceso en línea:https://hdl.handle.net/2117/167809
Access Level:acceso abierto
Palabra clave:High performance computing
Big data
Supercomputers
distributed systems
programming
programming models
framework
library
python
parallel programing
parallel
pycompss
compss
redis
storage
object
object management
hecuba
dataclay
hashing
hyperloglog
algorithms
programming languages
collections
arrays
java
Càlcul intensiu (Informàtica)
Dades massives
Supercomputadors
Àrees temàtiques de la UPC::Informàtica
Descripción
Sumario:Object management represents a substantial fraction of the total computing time in any distributed application, and it also adds complexity in terms of source code. This project proposes and implements a set of features aimed to improve both the usability and performance of distributed applications