Today, IBM announced its much anticipated POWER9 chip which will generally be available starting today.
While the chip’s architecture can do a lot more like databases, it excels at Artificial Intelligence and machine learning.
In a world that requires increasing amounts of computing power to handle the resource-intensive demands of workloads like artificial intelligence and machine learning, IBM had the foresight to see the need for an AI and ML optimized chip and systems, and the multinational technology company delivered it.
POWER9 has been a five-year journey focused on the next generation AI needs of real customers and because of that, some experts are predicting it receiving much greater customer adoption compared to the previous edition (POWER8) and are already chalking it up as a big win for IBM.
The company intends to sell the chips to cloud vendors including Google and to third-party manufacturers.
In addition to the POWER9 chip, IBM is releasing its first servers in two configurations with support for POWER9. The AC922 server is currently deployed and running inside Summit and Sierra supercomputers at CORAL, the Department of Energy collaboration between Oak Ridge, Argonne and Livermore Labs.
The AC922 server in combination with the POWER9 chip is a very fast machine learning and HPC platform.
IBM designed the POWER9 chip specifically to improve performance on common AI frameworks such as Chainer, TensorFlow, and Caffe. The company claims the new chip will increase workloads running on these frameworks by four folds.
This would allow data scientists building and running models on a POWER9-powered machine increased speed, which should allow them to run and complete model creation much faster.
Some experts believe IBM has really differentiated itself from the competition with the POWER9 chip.
“With its new systems architecture that is optimized for accelerators used in machine learning, IBM’s Power9 is literally the Swiss Army knife of ML acceleration as it supports an astronomical amount of IO and bandwidth, 10X of anything that’s out there today,” Patrick Moorhead, principal analyst at Moor Insights & Strategy was quoted as saying in a TechCrunch report.
The release of the processor is a continuation of IBM’s aggressive approach to capturing high-growth market segments like artificial intelligence.
IBM, by aligning their strategy across segments such as artificial intelligence, machine learning, and deep learning, will be able to better compete in hyperscale datacenter and broader market data center initiatives.
This strategy has the potential to positively impact direct revenue for IBM and enable new larger scale data center deployments.
Right out of the gate the POWER9 chip is generating business for IBM.
Summit, the supercomputer built by Lawrence Livermore and Oakridge national laboratories, will be powered by the chip. Livermore says the supercomputer will be built on top of thousands and thousands of the POWER9 computers at a cost of $325 million.