Selecting a Version of Edge Microvisor Toolkit for Demonstrating the Capabilities of Intel Platforms #521
stevenhoenisch
started this conversation in
General
Selecting a Version of Edge Microvisor Toolkit for Demonstrating the Capabilities of Intel Platforms
#521
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
The principal versions of Edge Microvisor Toolkit showcase the capabilities of Intel platforms for edge AI workloads through yet-to-be-upstreamed Linux patches from Intel that optimize performance and other capabilities for Intel processors.
The immutable and mutable versions of Edge Microvisor Toolkit -- including a standalone node prepared for partner evaluation and a real-time developer node designed with the Preempt RT Linux Kernel for predictable performance -- results in a reference Linux operating system primed to demonstrate how Intel processors can empower operating system vendors and other technology partners to optimize their platforms for running edge and AI solutions.
As part of Open Edge Platform, the toolkit's versions, which have different profiles and installation options, are available as open-source demonstration vehicles on GitHub:
Edge Microvisor Toolkit Standalone Node, an immutable version integrated with Kubernetes and foundational extensions to deploy and rapidly evaluate edge AI applications on virtual machines and containers powered by Intel platforms.
Edge Microvisor Toolkit Developer Node, a mutable, ready-made version for development, customization, testing, benchmarking, and VM exploration.
Edge Microvisor Toolkit with Preempt_RT, a version you build by using a predefined JSON file with a custom configuration for real-time computing. The Preempt_RT patch for Linux Kernel 6.12 improves real-time performance.
Here's a diagram that illustrates the main differences between the standalone node and the developer node:
Through its optimizations for Intel platforms, these versions showcase distinctive capabilities for operating system vendors and other partners, who can typically obtain early access to next-generation Intel platform innovations, features, and optimizations. The versions support the following Intel processors:
You can also build your own mutable or immutable version by selecting a predefined JSON image configuration file and a package list file that determine the image's properties and capabilities. In addition, you can create your own image configuration file and package list to build exactly the version that you want -- a powerful approach to customizing a Linux operating system to validate your edge AI workload on Intel processors.
Another option is use the mutable, immutable, and real-time versions available for Edge Manageability Framework to automate the rollout of new OS profiles to compatible hardware platforms.
Evaluating Edge AI Applications on Intel Platforms with the Standalone Node
The standalone node is an immutable version of Edge Microvisor Toolkit integrated with Kubernetes and foundational extensions to deploy and rapidly evaluate edge AI applications on virtual machines and containers powered by Intel platforms. To deliver compatibility and optimal performance for various workloads, including AI/ML.
To ease evaluation, the Edge Microvisor Toolkit Standalone Node is designed to give you critical insights into the capabilities of Edge AI applications for various use cases and scale-out scenarios.
To begin the evaluation process, you download the installer on a laptop or development system and then create a bootable USB, which includes all the software components to install the Edge Microvisor Toolkit, Kubernetes, foundational extensions, and the Kubernetes Dashboard. After the edge node is running, you can evaluate various Edge AI applications, pipelines, and microservices from the Intel Edge services as well as open-source repositories by using standard tools like Helm. See the installation instructions on GitHub.
Building a Custom Image with the Developer Node
With the developer node, you can create a custom mutable or immutabe image of Edge Microvisor Toolkit for development and testing. When you download the mutable host ISO image, it contains essential pre-installed packages for a ready-to-use base environment, but you can install additional RPM packages tailored to your needs. See the installation instructions.
Alternatively, you can use a flexible toolchain to build or customize your own mutable or immutable images. Building a custom image can help you identify the operating system parameters and packages that work best for your requirements and workloads:
Deploying the Toolkit with Edge Manageability Framework
The versions of Edge Microvisor Toolkit that are coupled with Edge Manageability Framework help deliver early access to next-generation Intel platform features. Edge Manageability Framework automates the rollout of new OS profiles to compatible hardware platforms so that you can quickly deploy the latest Intel optimizations. For more information, see Deployment with Edge Manageability Framework and the project's GitHub repository.
Participating in the Project
Vendors, developers, and technology partners can take part in the GitHub community for the solution in various ways: contributing code, proposing a design, downloading and trying out a release, opening an issue, benchmarking application performance, and participating in Discussions. Check out the following resources:
Beta Was this translation helpful? Give feedback.
All reactions