How Active Compute Fabric Improves AI Workload Efficiency
Cornelis, a startup focused on improving how AI chips communicate, announced on Monday that it raised $205 million in a funding round led by IAG Capital Partners. The company also unveiled a new product called Active Compute Fabric, designed to reduce delays in data transfer between graphics processing units. This investment aims to strengthen Cornelis’s position in the competitive AI infrastructure market.
Breaking news
Canonical’s Web Presence Hit by Second DDoS Assault This Year
Loops AI Secures $3.5 Million to Power a New AI Operating System for E‑Commerce
Winverse AI Secures Tens of Millions in Seed Funding from DeepSeek and Long‑Z
Google Adds Pixel 9 Series to Certified Refurbished StoreThe funding will support the development and deployment of Active Compute Fabric, which addresses inefficiencies where GPUs spend significant time idle while waiting for data. By optimizing network performance between AI processors, the technology seeks to maximize computational utilization. Cornelis argues that current systems waste valuable processing cycles due to bottlenecks in data movement, a challenge that grows more pronounced as AI models increase in size and complexity.
Can Cornelis Disrupt Established Players in AI Hardware?
Active Compute Fabric functions as a specialized network layer that prioritizes low-latency communication between chips in AI servers. Unlike traditional networking approaches, it is engineered specifically for the demands of machine learning workloads, where rapid data exchange is critical. Cornelis claims the technology can reduce idle time for GPUs, thereby increasing the effective throughput of AI training and inference tasks. The product is intended for integration into existing AI server architectures without requiring major redesigns.
Despite Nvidia’s dominant position in AI chips and related systems, Cornelis believes there is room for innovation in the interconnect space, which has seen less competition. The company points to growing demand for scalable, efficient AI infrastructure as cloud providers and enterprises expand their AI capabilities. By focusing solely on networking enhancements rather than chip design, Cornelis aims to complement rather than directly replace existing GPU vendors. Analysts note that success will depend on real-world performance gains and adoption by major hardware manufacturers.
What is Active Compute Fabric and what problem does it solve? Active Compute Fabric is a networking technology developed by Cornelis to minimize delays in data transfer between AI chips. It targets the issue of GPU idle time caused by waiting for data, aiming to improve overall processing efficiency.
Frequently Asked Questions
Who led the funding round for Cornelis and how much was raised? IAG Capital Partners led the $205 million funding round for Cornelis, announced on Monday. The investment will support product development and market expansion.
Why is Cornelis focusing on networking rather than chip design? Cornelis believes that optimizing communication between existing AI chips offers a significant opportunity to boost performance without requiring new hardware. This approach allows the company to work alongside established chipmakers like Nvidia rather than compete directly.
