AI infrastructure faces a critical physical barrier: managing the immense heat and electricity demanded by ever-larger systems. Akash Systems is addressing this by applying diamond-based thermal materials directly to server GPUs, including those from Nvidia. This innovation lowers temperatures, allowing data centers to extract more compute power from existing energy supplies.
Felix Ejeckam, CEO of Akash Systems, explained that the company has "solved the heat problem in data centers and AI." He noted that limited energy supply and inefficient usage are major challenges. Akash Systems utilizes synthetic diamond, the world's most thermally conductive material, to enable efficient energy use for compute needs. Applying diamond to GPUs can reduce chip temperatures by 10 to 15 degrees Celsius.
This temperature reduction offers significant operational savings and unlocks additional compute capacity. By decreasing the power required for cooling, data center operators can redirect energy to run more workloads or achieve substantial cost savings. Ejeckam stated that their technology can provide operators with approximately one million dollars per server in value.
With power availability becoming a constraint in many regions, efficiency-driven technologies are crucial for scaling AI operations. Akash Systems is integrating its diamond-cooled servers with GPUs from vendors like Nvidia and AMD. Ejeckam emphasized that the compute problem is fundamentally an energy problem, and their solution helps double capacity within existing infrastructure rather than requiring new power plants.