— Dynamically scales clocks, based on characteristics of the workload, for maximum application performance. This ensures that each application runs at the highest clocks while remaining within the power and thermal envelope.
Double shared memory and register file
— Increase effective bandwidth with 2x shared memory and 2x register file compared to the Tesla K20X and K10.
— Increase data center energy efficiency by powering down idle GPUs when running legacy
— Efficiently and easily schedule MPI ranks across GPUs, increasing GPU utilization and ease of programming.
— Manage GPU processors in computing systems with widely used cluster/Grid solutions.
— Error Correcting Codes (ECC) memory protection for both internal memories and external GDDR5 DRAM meets a critical requirement for computing accuracy and reliability in supercomputing and data centers.