Introduces cache-aware scheduling for AI workloads alongside filesystem improvements and expanded hardware support for Nvidia, AMD, Intel and Apple silicon.
Linux 7.2 introduces cache-aware scheduling designed for AI workloads, alongside filesystem improvements and broader hardware support. The scheduling approach takes CPU cache hierarchies into account to reduce memory-related bottlenecks in compute-intensive workloads such as AI training and inference.
The cache-aware scheduling approach is intended to improve resource utilisation when multiple workloads share the same infrastructure. By taking CPU cache hierarchy into account, the scheduler aims to reduce interference between workloads and improve efficiency.
The release also expands hardware support across processor and graphics platforms. Nvidia receives improved GPU driver support, while AMD gains support for newer EPYC processors. Intel receives optimisations for newer architectures, while Apple silicon receives improved support in virtualisation environments, including for M-series chips.
Storage also receives filesystem and I/O improvements aimed at database and containerised workloads. Linux 7.2 also expands support for laptop hardware through improvements to power management, touchpad drivers and Wi-Fi, helping improve compatibility for systems running Linux natively.
The cache-aware scheduling changes, combined with filesystem and hardware updates, extend Linux 7.2’s support for AI workloads, cloud environments and developer systems.
