Best Practices for Architecting Performance and Capacity in the Burst Buffer Era
SessionBurst Buffers and GPU's
Session ChairTrey Breckenridge
Presenter
Event Type
Exhibitor Forum
Accelerators
Architectures
I/O
Storage
Location155-F
DescriptionNow that Burst Buffers are commercially available, how will this game-changing technology disrupt the way high performance compute, file and storage systems are architected?
Incorporating a Burst Buffer inherently changes how you will select, procure and aggregate compute/storage/networking components in order to achieve performance and capacity goals. The positive impact is significant:
- Performance and capacity that have been historically intertwined are finally bifurcated, now allowing you to scale each independently
- Significant I/O, application acceleration and faster time to results can be realized
- 20% or more computational time can be unlocked unleashing significant ROI
- I/O bottlenecks and I/O bound applications that bring parallel file systems to their knees can be resolved without code modifications
- Footprint and power planning metrics can now be substantially reduced
- Exascale planning metrics can be modeled in greater, more accurate detail
Incorporating a Burst Buffer inherently changes how you will select, procure and aggregate compute/storage/networking components in order to achieve performance and capacity goals. The positive impact is significant:
- Performance and capacity that have been historically intertwined are finally bifurcated, now allowing you to scale each independently
- Significant I/O, application acceleration and faster time to results can be realized
- 20% or more computational time can be unlocked unleashing significant ROI
- I/O bottlenecks and I/O bound applications that bring parallel file systems to their knees can be resolved without code modifications
- Footprint and power planning metrics can now be substantially reduced
- Exascale planning metrics can be modeled in greater, more accurate detail
Presenter








