NVIDIA Makes Waves in the Industry with ADLINK

May 17, 2021 by Jake Hertz

ADLink has tapped into NVIDIA’s resources for two new products this month, sealing its impact in the electronics industry.

In many respects, NVIDIA is one company that is producing some of the best hardware computing resources in the industry. Between its GPU offerings, AI architectures, and its new data center CPU, much of NVIDIA hardware comes highly lauded and highly utilized. 


NVIDIA Grace is the company’s first data center CPU.

NVIDIA Grace is the company’s first data center CPU. Image used courtesy of NVIDIA 


One company that has been leveraging its AI resources to the fullest extent is ADLINK. This month alone, ADLINK has released two new products in the AI space based on NVIDIA hardware – an MXM graphic card and an industrial smart camera. 

This article will look at both releases to get a feel for how companies are benefiting from NVIDIA’s offerings. 



The first of ADLINK’s newest AI offerings came last week when it announced the release of the industry’s first embedded MXM-based graphics modules based on NVIDIA’s Turing architecture. Designed specifically with edge AI acceleration in mind, the MXM modules offer a nice blend of power, size, and performance.


New MXM offerings from ADLINK.

New MXM offerings from ADLINK. Image [modified] used courtesy of ADLINK


All in all, the new family consists of three models, ranging in memory (128-bit, 192-bit, or 256-bit GDDR6 memory) and power (50 W, 80 W, or 110 W TCP). 

According to the datasheet for the EGX-MXM-T1000, the device offers 896 CUDA cores, 4 GB GDDR6 memory, and a max memory bandwidth of 192 GB/s maximal memory bandwidth. This infrastructure allows the device to achieve a 2.6 TFLOPS peak FP32 performance while offering a power consumption that can be as low as 50 W. With a form factor of 82 x 70 mm, its new MXM cards are about ⅕ the size of a standard PCIe card. 


ADLINK, NVIDIA, and Smart Cameras 

The second release from ADLINK is its newest industrial AI smart camera, which it claims is the industry’s first based on the NVIDIA Jetson platform

The Jetson Xavier NX module is NVIDIA’s edge AI System on Module (SoM), consisting of 384 CUDA cores, 48 Tensor Cores, 6 Carmel ARM CPUs, and two NVIDIA Deep Learning Accelerators (NVDLA) engines. The SoM can achieve up to 21 TOPS at best and up to 14 TOPS at a low 10 W operation for battery-powered systems.


NVIDIA’s Jetson Xavier NX SoM.

NVIDIA’s Jetson Xavier NX SoM. Image used courtesy of NVIDIA


These features make the Jetson Xavier NX a perfect fit for ADLINK’s smart camera application. Leveraging this high level of integration and performance, ADLink’s new camera, the NEON-2000-JNX series, comes fully suited for the rapid development and deployment of smart camera applications. 

The camera comes with the ability to support six different camera configurations and offers two MIPI image sensors along with an embedded image signal processor. The company claims that the high level of integration also allows for high resistance to reliability issues, including thermal and EMC. 


NVIDIA’s Industry Impact 

Recently, NVIDIA has been rising to prominence in the computing industry, with its impact being seen far and wide.

Engineers anywhere from data centers to edge computing could have benefited from its GPU offerings, and now, in an ongoing conversation about modular design, their Jetson SoMs are finding prominence in the marketplace as well. 

Companies like ADLINK have leveraged these resources to offer new and impressive end products to their users. If recent history is any indicator, we can expect to see more NVIDIA-based products, from both ADLINK and elsewhere, reach the market in the near future.



Interested in more news and innovations from NVIDIA? Find out more down below.

Is the NVIDIA-Arm Deal Dead in the Water?

NVIDIA Scales the Cloud With a Reimagined Data Processing Unit (DPU)

Keeping Power Integrity a Priority With the Right Capacitor