Power servers

Your onramp to IBM PowerAI

Share this post:

As offering managers on IBM’s PowerAI team, my colleagues and I assist worldwide clients, Business Partners, sellers and systems integrators daily to accelerate their journey to adopt machine learning (ML), deep learning (DL) and other AI applications, specifically based on IBM PowerAI and PowerAI Vision.

Thanks to our collective efforts we been able to establish an adoption roadmap for our PowerAI offering which we’d like to pass along to our future clients and partners. This fine-tuned roadmap is based on years of experience and has proven useful in hundreds of client engagements. An additional benefit is the summarization of the processes we recommend into a small number of high-value steps. These 5 steps, once actualized with factors local to each client’s needs, will accelerate your journey to trying AI and assessing its benefits in your environment, with actions in place to reduce risk and costs:

  1. Provide important background info: First, make sure all members of your staff involved in future AI projects are offered the means to learn more about AI, ML and DL. We recommend starting here, however, you should feel free to leverage the vast amount of additional information that is publicly available.
  2. Identify priorities: Next, working closely across a team of both business and technical leaders, identify top-priority business problems or growth opportunities that your client or company views as potential candidates for exploring the potential benefits of AI projects or machine learning/deep learning solutions. During this phase of the adoption effort, it may be very beneficial to augment your current staff or in-house skill base by bringing in additional subject matter experts in AI/ML/DL, sourcing them from hardware and software vendors that offer AI solutions, consulting firms with AI practices, or systems integrators.
  3. Select data sources: Identify the wide variety of data sources, including “Big Data”, that you can and will leverage when applying ML/DL algorithms to solve these current business problems your team has identified.
  4. Find the right software/hardware: Select and gain access to machine learning or deep learning software along with accelerated hardware so that you can start your AI project development, tests and proof of concept (POC) process. An effective POC journey will leverage your data along with your success criteria to help ensure that AI projects can move forward at full speed and expand rapidly at scale. Build and train the ML/DL models and then demonstrate where the AI POC effort was successful. Other areas may either need newer approaches or more work to yield the required incremental business benefits from AI.
  5. Create the Production Environment: Plan out your production machine learning and/or deep learning deployment environment. Now would be a good time to establish a complete reference architecture for your production AI systems, and you can start identifying and planning to change any in-house business processes or workflows as a result of better-leveraged and newer AI algorithms, big data and the next generation of accelerated IT infrastructure.

Finally, as part of your onramp process, we recommend you learn more about IBM’s PowerAI. A good starting point to begin your journey is here. Feel free to reach out to us for more information and to leverage more detailed guidance. As a reference point you may also want take a second to review the top reasons that hundreds of clients have already chosen to run IBM’s PowerAI:

top reasons to choose PowerAI

More Power servers stories

Top IBM Power Systems myths: “IBM AIX is dead and Unix isn’t relevant in today’s market” (part 2)

AI, IBM Systems Lab Services, Power Systems

In part 1 of this series, we started to look at the myth that IBM AIX and Unix are no longer relevant. We talked about the Unix wars that began in the 1980s and how the market has evolved since then. Now, let’s consider the evolution of AIX in the past few decades and the ...read more


A conversation about metadata management featuring Forrester Analyst Michele Goetz

AI, Big data & analytics, Storage

I recently had an opportunity to speak with Forrester Principal Analyst Michele Goetz following a research study on the subject of metadata management conducted by Forrester Consulting and commissioned by IBM. Michele’s research covers artificial intelligence technologies and consultancies, semantic technology, data management strategy, data governance and data integration, and includes within its scope the ...read more


IBM i – The driverless variant of IT infrastructure

Big data & analytics, Modern data platforms, Power servers

I am, quite frankly, not looking forward to the advent of autonomous automobiles. I happily drive a sporty, stick-shift vehicle myself. But what does intrigue me is autonomous IT infrastructure. I keep track of developments in autonomous IT and consider it to be an early-stage trend that will not reach maturity for several more years. ...read more