Inspur is expanding its AI (artificial intelligence) computing portfolio from the cloud data center to edge computing, with the launch of new AI server. Called Inspur NE5250M5, the new AI server will feature two NVIDIA V100 Tensor Core GPUs or six NVIDIA T4 GPUs.
Inspur NE5250M5 will be useful for most of the compute-intensive AI applications. These applications can include autonomous vehicles, smart cities, and smart homes. Moreover, it will also enable 5G edge applications like internet of things (IoT), MEC and NFV.
To enable these applications, the new AI server will feature a telecom server form factor. This form factor will be 2U tall, 430mm deep, and 19 inches wide. Since the design of new server is about half the depth of the traditional standard server, it can be directly installed with telecom servers on the CT data center.
“This NE5250M5 is the latest example of Inspur’s commitment to providing a diverse array of cutting-edge AI computing power—from the cloud data center to edge computing,” said Jun Liu, GM of Inspur AI & HPC.
“The dynamic nature and rapid expansion of AI workloads require an adaptive and optimized set of hardware, software and services for developers to utilize as they build their own solutions. With the addition of the NE5250M5, the Inspur AI portfolio continues to expand to provide the industry’s broadest, deepest selection of AI products and solutions.”
Also read: Nvidia acquires Israeli networking firm Mellanox for $6.9B
Inspur NE5250M5 will come with several options for mounting. Customers will be able to install it on walls or choose to use space-constrained edge environments. While designing it, the company has taken care that the new server can reliably work regardless of heat, dust, corrosion, electromagnetism, and shock.
