Accelerated servers

Enterprise AI requires a new breed of servers. Servers that enable the cutting-edge AI innovation data scientists desire, with dependability IT requires.

IBM Power Systems offers the fastest way to deploy accelerated databases and deep learning frameworks – with enterprise-class support.

IBM Power System AC922

IBM Power Systems Accelerated Compute Series (AC922) is the best server for enterprises looking to deploy AI. The AC922 is the fastest path to and for acceleration, unleashing accelerated computing potential, in the post CPU-only era.

IBM Power System S822LC for High Performance Computing

Provides innovation from the OpenPOWER Foundation ecosystem, with up to two NVIDIA® Tesla® GPU Accelerators and Mellanox® InfiniBand to deliver faster insights and flexible POWER8 architecture.

IBM Power System S824L (with NVIDIA technology)

Leverages OpenPOWER Foundation technology to accelerate HPC and technical computing applications. Utilize the strength of POWER8 with NVIDIA® GPUs and offload parallel operations for up to two NVIDIA Tesla® GPU Accelerators.

Starting at $16,419.00 (IBM US Web price)*


* Price does not include tax and is subject to change without notice. Price indicated applies to specific model which may not have all available features and functions. Different configurations have higher prices. Operating system is available for an additional charge.

How businesses benefit from accelerated servers

Hitting the wall with server infrastructure?

Businesses around the world are responding vigorously to the new opportunities offered by AI. These workloads require advanced infrastructures that many organizations are struggling to identify and build. They’re hitting a wall with their existing infrastructure, and need to find a way forward to maintain competitive advantage.

CORAL Summit supercomputer: likely the world's most powerful technology

The compute performance behind CORAL Summit supercomputer is now available to the enterprise. Don’t believe us?

See a sneak peek of what 200+ petaflops of high-performance computing and 3 exaflops of AI-as-a-service looks like.