Geomancy: Automated Performance Enhancement through Data Layout Optimization

The size and complexity of large storage systems, such as high-performance computing (HPC) systems, inhibit rapid effective restructuring of data layouts to maintain performance as workloads shift. To address this issue, we have developed Geomancy, a tool that models the placement of data within a d...

Descripción completa

Detalles Bibliográficos
Autores principales: Bel, Oceane, Chang, Kenneth, Tallent, Nathan R., Duellmann, Dirk, Miller, Ethan L., Nawab, Faisal, Long, Darrell D. E.
Publicado: 2020
Acceso en línea:https://dx.doi.org/10.1109/ISPASS48437.2020.00025
http://cds.cern.ch/record/2799930
Descripción
Sumario:The size and complexity of large storage systems, such as high-performance computing (HPC) systems, inhibit rapid effective restructuring of data layouts to maintain performance as workloads shift. To address this issue, we have developed Geomancy, a tool that models the placement of data within a distributed storage system and reacts to drops in performance. Our approach to optimizing throughput offers benefits for storage systems such as avoiding potential bottlenecks and increasing overall I/O throughput from 11% to 30%.

Ejemplares similares