Implementing the OpenSSF Best Practices Badges & Scorecards into Your Project

Christopher Robinson (CRob), Intel & David A. Wheeler, Linux* Foundation

Live

Tuesday, September 19, 14:30-15:10 GMT+2

Energy Proportional Edge Computing Infrastructure

Francesc Guim, Intel

Live

Wednesday, September 20, 16:50-17:30 GMT+2

Sponsored Session—The Search for Transparency and Accountability in the Age of AI: Responsible AI (RAI) & Explainable AI (XAI) as Essential Tools

Ezequiel Lanza, Intel

Live

Thursday, September 21, 11:00-11:40 GMT+2

Intel® Energy Estimator

Intel® Energy Estimator is a tool that is designed to estimate the energy consumption of workloads running on Intel CPUs. The Intel Energy Estimator client library acts as an "energy calculator," offering consumption estimates based on key metrics like CPU use. This fills a crucial gap in energy measurement, covering virtualized and cloud environments, on-premises setups, and bare-metal scenarios.

Intel Energy Estimator estimates rely on Intel's power characteristics, ensuring access to the most up-to-date and accurate data. Thanks to its versatile nature as a library, Intel Energy Estimator can be applied across a range of use cases and applications, including web apps, dashboards, monitoring and alerting systems, and various scripts.

Efficient AI: Empower Large Language Models with Intel® Extension for PyTorch* to Combat Carbon Emissions

This demo showcases how Intel® Extension for PyTorch* can reduce carbon emissions while maintaining the performance of large language models (LLMs). With just a few lines of code, you will see the improvements that can be achieved in latency and power consumption. Come watch our demo to learn more.

Accelerate the Design and Development of Vision-Enabled Self-Checkout Solutions

The use of camera vision has been transforming many industries, and retail applications are no different. Automated self-checkout solutions are complex and require a good understanding of hardware and software to build solutions that can be scaled.

The Automated Self-Checkout Retail Reference Implementation is an open source initiative that provides critical software that is highly optimized to run on Intel hardware. It helps developers bootstrap their self-checkout implementation and accelerate their development.

This demo showcases the various software-optimized code that developers can use while removing anticipated pitfalls developers encounter when stringing together multiple software components. Get a highlight of typical hardware resource issues from compute to memory and the choices developers can make when running AI-powered workloads across CPU, iGPU, and discrete GPUs.