Sentia's Power AI Innovation Centre
Helping to accelerate HPC, Analytics & Artificial Intelligence
As data continues to explode, our ability to derive value from it, is the new business challenge.
Cognitive Computing unlocks all forms of data, and enables businesses to make better decisions, optimize processes, and truly understand their customers' needs.
IBM's Power AI platform helps accelerate the journey to knowledge by packaging a collection of open-source frameworks, software and libraries, making it easy to acquire, install and achieve a deep learning infrastructure.
Customized for your needs
Sentia's Power Systems experts will customize our IBM Power Accelerated Computer Server (AC922) environment so you can comfortable test drive it and its robust deep learning framworks.
IT Infrastructure Redesigned for Enterprise AI
AC922 delivers unprecedented performance for modern HPC, analytics, and artificial intelligence. Enterprises can now deploy data-intensive workloads, like deep learning frameworks and accelerated databases, with confidence.
AC922 enables the cutting-edge AI innovation data scientists desire, with the dependability IT requires.
Capable of tackling the world's biggest AI challenges
Faster I/O - up to 5.6x more I/O bandwidth then x86 servers
The AC922 includes a variety of next-generation I/O architectures, providing 2 to 5.6x the bandwidth over x86 servers.
The best GPUs - 2 - 6 NVIDIA Tesla V100 GPUs with NVLink
The AC922 pairs POWER9 CPUs and NVIDIA Tesla V100 with NVLink GPUs. This is the only server capable of delivering the I/O performance between CPUs and GPUs.
Extraordinary CPUs - 2x POWER9 CPUs, designed for AI
POWER9 supports up to 5.6x more I/O and 2x more threads than its x86 contemporaries. The POWER9 is available on configurations with anywhere between 16 and up to 44 cores.
Next Gen PCle - PCle Gen4 2xfaster vs PCle Gen3 in x86
The AC922 is the industry's first server to feature 2x the data bandwidth of PCle Gen3 interconnects found in x86 servers.
Simplest AI Architecture - Share RAM across CPUs and GPUs
AI models easily outgrow GPU memory capacity in most x86 servers. With the AC922, CPU to GPU coherence is 5.6x faster using NVLink. Sharing memory between CPUs and GPUs avoids the PCle3 bottleneck found in today's x86 servers.
Enterprise-ready - Power AI DL frameworks with IBM support
PowerAI DL framework simplifies deep-learning deployment and performance. It unlocks a new, simpler, end to end toolchain for AI users.