Nvidia GRID

Nvidia GRID was a family of graphics processing units (GPUs) made by Nvidia, introduced in 2008, that were targeted specifically towards GPU virtualization and cloud gaming.^[1] The Nvidia GRID includes both graphics processing and video encoding into a single device which is able to decrease the input to display latency of cloud based video game streaming.^[2] It was previously used by Nvidia's GeForce Now, a paid cloud gaming service.

This GRID K1 GPU provides VDI for four seats using four independent GK107 GPUs with 4 GB of graphics memory each.

While many of Nvidia’s cards are known for gaming, there has been a recent growth of business applications that are GPU-accelerated.^[timeframe?] The Nvidia GRID K1 and K2 are being integrated with Supermicro server clusters for use with 3D-intensive applications such as graphics and computer aided design (CAD).^[3] In 2015, Microsoft began including Nvidia GRID as part of its Azure Enterprise cloud platform targeted towards professionals such as engineers, designers and researchers.^[4]

Specifications^[5][6][7]

	GRID K1	Grid K2
Microarchitecture	Kepler
Number of GPUs	4× GK107	2× GK104
Number of CUDA cores	4× 192	2× 1536
Memory site	4× 4 GB DDR3	2× 4 GB GDDR5
Max power	130 W	225 W