On Aug. 24, Intel held the first in a series of three Chalk Talk sessions to provide context on the company's journey to build and deliver the silicon, software and platforms customers need to succeed in today’s digital era.
This first session focused on Intel's broad portfolio of integrated and discrete accelerators. Leaders from across Intel’s business units shared how the company solves critical workloads for customers across industries and verticals. Intel executives who participated in the acceleration Chalk Talk included:
- Sailesh Kottapalli, chief data center CPU architect, spoke on Intel® Xeon® architecture and data decompression.
- Kavitha Prasad, vice president and general manager of Datacenter, AI and Cloud Execution and Strategy, spoke on AI acceleration with Intel software and hardware.
- Sachin Katti, chief technology officer in the Network and Edge Group, spoke on acceleration for networking.
- Jeff McVeigh, vice president and general manager of the Super Compute Group, spoke on acceleration for high performance computing.
“By taking a workload-first, interactive approach, the Chalk Talk series is meant to unpack the ‘why’ behind the bets we’re making with our strategy and highlight the real-world impact our technology innovations are having on customers across industries,” said Sandra Rivera, executive vice president and general manager of the Datacenter and AI Group at Intel . “We realize there is no one-size-fits-all with compute – and our customers are transforming their businesses, leveraging the capabilities of heterogeneous computing environments."
Intel’s next two Chalk Talks will focus on Intel's approach to security in September and Sapphire Rapids processing and packaging in October. Future Chalk Talk replays will be published on the Intel Newsroom after those events.
All the Chalk Talks:
Editor's Note: The video was edited Sept. 13, 2022, to remove a reference to the Habana® Gaudi® AI training processor delivering 40 times better price performance. The processor delivers up to 40% better price performance than comparable Nvidia GPU-based training instances.