NVLink interconnect

E758260

GPU interconnect NVIDIA technology high-speed interconnect technology

NVLink interconnect is NVIDIA’s high-speed GPU-to-GPU and GPU-to-CPU communication technology designed to provide significantly greater bandwidth and lower latency than traditional PCIe connections in high-performance computing and data center systems.

Try in SPARQL Jump to: Surface forms Disambiguation Statements Elicitation Referenced by

All labels observed (12)

Label	Occurrences
NVLink	4
NVIDIA NVLink	3
NVLink 2.0	2
NVIDIA NVLink (on selected models)	1
NVIDIA NVLink (via compatible systems, if applicable)	1
NVIDIA NVSwitch-based systems	1
NVLink 1.0	1
NVLink 3.0	1
NVLink 4.0	1
NVLink interconnect canonical	1
NVLink interface	1
NVSwitch	1

How this entity was disambiguated

This entity first appeared as the object of triple T8822664 — resolving that mention is where its identity was fixed. The disambiguator weighed these candidate entities and picked the highlighted one (or “None”, minting a new entity). This is how homonymy is resolved: the same surface form can point to different entities.

NED1 Entity disambiguation (via context triple) gpt-5-mini-2025-08-07

Target entity: NVLink interconnect
Context triple: [NVIDIA Tesla data center GPUs, supports, NVLink interconnect]

A. NVIDIA Ampere architecture
NVIDIA Ampere architecture is a GPU microarchitecture from NVIDIA that powers RTX 30-series graphics cards, delivering significant improvements in ray tracing, AI performance, and power efficiency over previous generations.
B. NVIDIA Ada Lovelace architecture
NVIDIA Ada Lovelace architecture is a GPU microarchitecture from NVIDIA that powers the RTX 40-series graphics cards, delivering major advances in ray tracing, AI acceleration, and power efficiency over previous generations.
C. Intel Omni-Path Architecture
Intel Omni-Path Architecture is a high-performance computing interconnect technology developed by Intel to provide low-latency, high-bandwidth communication for large-scale supercomputing and data center clusters.
D. GPUDirect
GPUDirect is an NVIDIA technology that enables high-speed, low-latency data transfers directly between GPUs and other devices or memory, bypassing the CPU to improve performance in data-intensive applications.
E. InfiniBand
InfiniBand is a high-speed, low-latency communications standard commonly used as the interconnect fabric in supercomputers and large-scale data centers.
F. None of above. chosen
G. Unsure - the case is ambiguous/there is not enough information to decide.

NED2 Entity disambiguation (via description) gpt-5-mini-2025-08-07

Target entity: NVLink interconnect
Target entity description: NVLink interconnect is NVIDIA’s high-speed GPU-to-GPU and GPU-to-CPU communication technology designed to provide significantly greater bandwidth and lower latency than traditional PCIe connections in high-performance computing and data center systems.

A. NVIDIA Ampere architecture
NVIDIA Ampere architecture is a GPU microarchitecture from NVIDIA that powers RTX 30-series graphics cards, delivering significant improvements in ray tracing, AI performance, and power efficiency over previous generations.
B. NVIDIA Ada Lovelace architecture
NVIDIA Ada Lovelace architecture is a GPU microarchitecture from NVIDIA that powers the RTX 40-series graphics cards, delivering major advances in ray tracing, AI acceleration, and power efficiency over previous generations.
C. Intel Omni-Path Architecture
Intel Omni-Path Architecture is a high-performance computing interconnect technology developed by Intel to provide low-latency, high-bandwidth communication for large-scale supercomputing and data center clusters.
D. GPUDirect
GPUDirect is an NVIDIA technology that enables high-speed, low-latency data transfers directly between GPUs and other devices or memory, bypassing the CPU to improve performance in data-intensive applications.
E. InfiniBand
InfiniBand is a high-speed, low-latency communications standard commonly used as the interconnect fabric in supercomputers and large-scale data centers.
F. None of above. chosen

Statements (47)

Predicate	Object
instanceOf	GPU interconnect ⓘ NVIDIA technology ⓘ high-speed interconnect technology ⓘ
architecture	serial high-speed links ⓘ
bandwidthAdvantageOver	PCIe 3.0 ⓘ PCIe 4.0 ⓘ
brand	NVIDIA NVLink ⓘ linked to: NVLink interconnect
comparedTo	PCI Express ⓘ
connects	GPUs to CPUs in supported architectures ⓘ multiple GPUs within a node ⓘ
designedFor	GPU-to-CPU communication ⓘ GPU-to-GPU communication ⓘ data center systems ⓘ high-performance computing ⓘ
developer	NVIDIA ⓘ linked to: NVIDIA Corporation
enables	high-bandwidth GPU memory sharing ⓘ scaling of multi-GPU systems ⓘ
generation	NVLink 1.0 ⓘ linked to: NVLink interconnect NVLink 2.0 ⓘ linked to: NVLink interconnect NVLink 3.0 ⓘ linked to: NVLink interconnect NVLink 4.0 ⓘ linked to: NVLink interconnect
hasFeature	cache-coherent communication in supported systems ⓘ high bandwidth ⓘ low latency ⓘ point-to-point links ⓘ
introducedBy	NVIDIA Pascal architecture ⓘ
latencyLowerThan	PCIe ⓘ linked to: PCI Express
manufacturer	NVIDIA ⓘ linked to: NVIDIA Corporation
marketSegment	cloud AI infrastructure ⓘ enterprise data centers ⓘ supercomputing ⓘ
relatedStandard	PCI Express ⓘ
supports	GPU peer-to-peer memory access ⓘ unified memory addressing in supported platforms ⓘ
supportsTopology	hypercube-like topologies ⓘ mesh topologies ⓘ
typicalUseCase	AI inference workloads ⓘ AI training workloads ⓘ large-scale data analytics ⓘ scientific simulations ⓘ
usedIn	NVIDIA A100 GPUs ⓘ linked to: NVIDIA A100 NVIDIA DGX systems ⓘ linked to: NVIDIA DGX NVIDIA H100 GPUs ⓘ linked to: NVIDIA H100 NVIDIA HGX platforms ⓘ NVIDIA NVSwitch-based systems ⓘ linked to: NVLink interconnect NVIDIA Tesla GPUs ⓘ linked to: NVIDIA Tesla data center GPUs
usedWith	NVSwitch ⓘ

How these facts were elicited