Product Advantages
-
Flexibility
Different G series servers provide the memory you need for graphics application, and P series servers are great for scientific computing or deep learning workloads.
Different G series servers provide the memory you need for graphics application, and P series servers are great for scientific computing or deep learning workloads.
-
Ecosystem
A comprehensive ecosystem provides support for multiple GPU applications and deep learning frameworks. G series servers support OpenGL and DirectX, while the P series servers support CUDA and OpenCL.
A comprehensive ecosystem provides support for multiple GPU applications and deep learning frameworks. G series servers support OpenGL and DirectX, while the P series servers support CUDA and OpenCL.
-
Ease of use
Obtain all sorts of graphic workstations, supercomputing applications, deep learning frameworks, and computing clusters in just a few clicks.
Obtain all sorts of graphic workstations, supercomputing applications, deep learning frameworks, and computing clusters in just a few clicks.
-
Cost effectiveness
Take advantage of pay-per-use or periodic subscriptions so you always have the latest hardware at your fingertips, with elastic scaling built right in.
Take advantage of pay-per-use or periodic subscriptions so you always have the latest hardware at your fingertips, with elastic scaling built right in.
Application Scenarios
Functions
-
HPC and AI
Superb single- and dual-precision computing power
Superb single- and dual-precision computing power
-
Video/Graphics Rendering
Professional video or graphics rendering
Professional video or graphics rendering
-
PI2PI2
Developed using NVIDIA Tesla T4 GPUs.
A single GPU offers 8.1 TeraFLOPS of single-precision computing and 130 TeraOPS INT8.
Uses 16 GB DDR6 GPU memory, 300 Gbit/s bandwidth, 1 NVENC encoder, and 2 NVDEC decoders.
Developed using NVIDIA Tesla T4 GPUs. A single GPU offers 8.1 TeraFLOPS of single-precision computing and 130 TeraOPS INT8. Uses 16 GB DDR6 GPU memory, 300 Gbit/s bandwidth, 1 NVENC encoder, and 2 NVDEC decoders.
-
Video RenderingVideo Rendering
G5 instances, developed using NVIDIA Tesla M60 GPUs and GPU pass-through, support 8/16 GB DDR GPU memory for heavy-load graphics design and video processing.
G5 instances, developed using NVIDIA Tesla M60 GPUs and GPU pass-through, support 8/16 GB DDR GPU memory for heavy-load graphics design and video processing.