Since the COVID-19 outbreak started in in January, we have seen many tech events cancelled or converted to virtual conferences. Form Mobile World Conference (MWC) being cancelled and Microsoft's Build 2020 heading online, tech events have taken a beating. Among them was NVIDIA's GPU Technology Conference (GTC), which went virtual.
The conference is now happening online and will run until March 26. One of the announcements made at GTC was Microsoft confirming it has expanded Azure Stack Edge with NVIDIA preview.
NVIDIA GPU preview first arrived at MWC 2019 and is a collaboration involving Microsoft Azure and NVIDIA's EGX platform. More specifically, Azure Stack Edge uses the NVIDIA T4 Tensor Core GPU to boost the capabilities of machine learning.
In its blog post, Microsoft says the preview has been adopted by industries who want to leverage Azure for machine learning.
Essentially, NVIDIA T4 Tensor Core GPU is leveraged by the Azure Stack Edge in order to bring a hardware boost to machine learning (ML) workloads. Microsoft states that since its introduction, it has observed a largely positive reception to the idea, with a variety of industries looking to take advantage of the improved Azure ML capabilities.
Benefits of Azure Edge Stack
- “Azure Machine Learning: Build and train your model in the cloud, then deploy it to the edge for FPGA or GPU-accelerated inferencing.
- Edge Compute: Run IoT, AI, and business applications in containers at your location. Use these to interact with your local systems, or to pre-process your data before it transfers to Azure.
- Cloud Storage Gateway: Automatically transfer data between the local appliance and your Azure Storage account. Azure Stack Edge caches the hottest data locally and speaks file and object protocols to your on-prem applications.
- Azure-managed appliance: Easily order and manage Azure Stack Edge from the Azure Portal. No initial capex fees; pay as you go, just like any other Azure service.”