publication . Article . 2017

Ultra-Performance Pascal GPU and NVLink Interconnect

Denis Foley; John Danskin;
Restricted
  • Published: 12 May 2017 Journal: IEEE Micro, volume 37, pages 7-17 (issn: 0272-1732, Copyright policy)
  • Publisher: Institute of Electrical and Electronics Engineers (IEEE)
Abstract
This article introduces Nvidia's high-performance Pascal GPU. GP100 features in-package high-bandwidth memory, support for efficient FP16 operations, unified memory, and instruction preemption, and incorporates Nvidia's NVLink I/O for high-bandwidth connections between GPUs and between GPUs and CPUs.
Subjects
ACM Computing Classification System: ComputingMethodologies_COMPUTERGRAPHICSComputerSystemsOrganization_PROCESSORARCHITECTURESSoftware_PROGRAMMINGTECHNIQUES
free text keywords: Shared memory, Uniform memory access, High Bandwidth Memory, CUDA Pinned memory, Flat memory model, Memory management, Parallel computing, Conventional memory, Interleaved memory, Computer science
Related Organizations
Powered by OpenAIRE Open Research Graph
Any information missing or wrong?Report an Issue