I don’t need to tell anyone how hot the topic of AI is today. We see it all the time in the press and industry commentary. I work with clients daily, and almost all are all keen to deploy AI in their business. While they understand the opportunities to leverage AI for operational efficiency and to create new business value, there are several questions they typically have. 

Many of the conversations I engage in focus on integrating AI into existing business workloads. As is the case with most enterprise projects, cost and time are always top of mind. How long will it take to get a return on my investment? How much is it going to cost me? How can I keep my AI data safe? The answer is, of course, “it depends”.

Part of the discussion around deployment focuses on what the right infrastructure for any individual AI workload is. The choice you make not only impacts cost, but how successful the result is in driving the expected increase in business value.

IBM LinuxONE, our enterprise Linux server, is an ideal infrastructure for many predictive AI workloads. Its strengths as a secure and data serving platform are well established. Many clients have moved proprietary or open-source databases to IBM LinuxONE to take advantage of leading capabilities for data privacy, security, and protection, as well as low cost of ownership for data serving at scale.

Now with IBM LinuxONE 4, you can take advantage of AI inferencing engineered for speed and scale with the security of IBM Hyper Protect. As a long-time innovator in systems design, IBM recognizes the importance of supplementing processing cores with dedicated resources to support key functions and collaborating with the open-source community. The LinuxONE 4 introduces an integrated AI accelerator, fine-tuned for low latency and high throughput performance. IBM’s internal benchmarks show that the LinuxONE Emperor 4 can efficiently handle up to 3.5 million inference requests per second, maintaining a rapid response time of just 1ms for tasks like Credit Card Fraud Detection¹.

This technology is not just applicable to credit card fraud protection. IBM, working with our clients, have identified use cases across industries that can take advantage of this technology to create new value for their business. IBM offers a no-charge AI on LinuxONE Discovery Workshop. This workshop is a great starting point and can help you evaluate potential use cases and define a project plan. If you would like to schedule a workshop on any of the AI use cases or products, email us at aionz@us.ibm.com.

LinuxONE 4 offers you an on-chip AI accelerator and the ability to add-on FIPS 140-2 Level 4 hardware. This enables IBM Hyper Protect Crypto Services and leverages confidential computing with IBM Hyper Protect Virtual Servers in addition to AI all on one platform. Leveraging AI with security helps keep your AI information secure while processing data.

Activating the IBM LinuxONE Ecosystem for client success

With its enhanced AI capability, IBM is opening more opportunities for IBM LinuxONE Ecosystem partners. Now, partners will be able to reach a broader range of clients and use cases. With ISVs (Independent Software Vendors) like OvationCXM, as well as CustomerXPs, and Exponential AI who will be speaking on our webcast, IBM is working to provide solutions for today’s AI and cybersecurity challenges.

IBM Hyper Protect Services (available on-prem and in the cloud) provides an environment that leverages the security of LinuxONE to protect sensitive data (often the case with AI use cases). Learn how startups (part of the IBM Hyper Protect Accelerator) like CheckUps, an medical services/AI app based in Africa, and Schwarzthal Tech, a financial crime/AI app based in Europe, are benefiting from this service: These benefits can be enabled to LinuxONE 4 using the underlying technology of a hardware security module (HSM) with FIPS 140-2 Level 4 protection for sensitive data.

You can experience machine learning on LinuxONE and improve your skills with IBM Z Xplore, a no-cost online learning platform with access to LinuxONE hands-on challenges.

To learn how LinuxONE can help you deliver new AI use cases for your business, join us on June 18 at 11 AM ET for our live webcast Unleash the power of AI with LinuxONE.

For more information about LinuxONE’s AI capabilities, visit the Integrated Accelerator for AI webpage to explore how IBM LinuxONE enables you to deploy AI inferencing at scale.

* DISCLAIMER: Performance result is extrapolated from IBM internal tests running local inference operations in an IBM LinuxONE Emperor 4 LPAR with 48 cores and 128 GB memory on Ubuntu 20.04 (SMT mode) using a synthetic credit card fraud detection model (https://github.com/IBM/ai-on-z-fraud-detection) exploiting the Integrated Accelerator for AI. The benchmark was running with 8 parallel threads each pinned to the first core of a different chip. The lscpu command was used to identify the core-chip topology. A batch size of 128 inference operations was used. Results may vary.