Engineered for AI inference
Accelerated, Power IC922 supports up to six NVIDIA® T4 Tensor Core GPU accelerators.
Advanced I/O interconnects
Power IC922 provides you with advanced interconnects (PCIe Gen4, OpenCAPI) to support faster data throughput and decreased latency. PCIe generation 4 delivers approximately 2x the data bandwidth of the PCIe generation 3 interconnect found in x86 servers.
Optimized AI Software
The Power IC922 inference server uses the same AI software (PowerAI Vision) designed for POWER9 across your Power AC922 and Power IC922.
The storage-rich Power IC922 has a max peak memory bandwidth of 170GB/s per socket, coupled with up to 24 SAS/SATA drives in a 2U system.
Modular and scalable
Power IC922 is built to be scalable, allowing you to scale your infrastructure to meet your business needs whether on-premises or in a private cloud environment, all with enterprise-class support.