IBM Corp. is all-in on artificial intelligence and machine learning. The company provides a broad range of cognitive products for businesses of all sizes. However, it is also using its other product lines, such as its storage business, to assist customers in developing products around Internet of Things and machine learning.

“We provide the storage platform with our flash technology to SparkCognition, a professional software company. They are a hot startup [with] a number of different use cases, including cybersecurity, real-time IoT and predictive analytics,” said Eric Herzog (pictured, left), vice president of product marketing and management, storage, and software-defined infrastructure, at IBM.

Herzog and Mark Godard (pictured, right), manager of customer success and partnerships at SparkCognition Inc., spoke to John Furrier (@furrier) and Dave Vellante (@dvellante), co-hosts of theCUBE, SiliconANGLE Media’s mobile live streaming studio, during IBM Interconnect 2017 in Las Vegas, Nevada. (*Disclosure below.)

During the interview, they discussed their unique partnership.

Using machine learning to predict and secure

SparkCognition is a three-year-old startup based in Austin, Texas, that developed patented AI-powered products that use an entirely cognitive approach to anti-malware, cybersecurity, and industrial safety and efficiency.

“We use data science, machine learning and natural language processing [in our] three products. SparkSecure is our predictive analytics/predictive maintenance product. SparkSecure is our network logs security product, and DeepArmor is a machine-learning endpoint protection product,” Godard said.

IBM’s flash storage supports companies with similar goals in using machine learning. Herzog noted that in each case the issues were the same. Customers need availability, and the storage can never fail because it supports real-time technology. Additionally, he mentioned that storage needs to be fast and resilient and it has to meet service-level agreements.

SparkCognition is spreading its wings in the Industrial Internet of Things, which the industry anticipates being an $80-billion market within five years, Godard stated. The company has advanced use cases in the industrial sector, working with a wind turbine company and oil and gas companies.

“Predictive maintenance is leading the way as the most mature application … [it] is valuable to clients because if you predict failures, you can optimize resources. Our goal is to be able to reduce maintenance costs and extend the useful life of assets,” said Godard.

SparkCognition is using a variety of platforms to support its products. Depending on the client or industry, the company can provide its AI and machine learning model software on-premise or in the cloud. These are very intensive applications that require and massive amounts of CPU, input/output and fast storage. According to Godard, it is critical to invest in premium hardware to get value from data quickly so it is useful and actionable.

Partners with benefits

IBM software-defined storage has allowed the software provider to deploy a small-scale implementation without having to own dedicated hardware. SparkConnect can spin up instances quickly to provision that small-scale deployment. Leveraging a software-defined approach offers a lower cost to its clients, Herzog said.

Both companies benefit from the partnership. They support each other in the ecosystem, and IBM provides SparkConnect access to Watson and other technologies that benefit the company’s efforts in IIOT.

Godard believes that IBM products are superior in many ways. And while SparkConnect is hardware agnostic, it often advises clients to run their products on IBM solutions to garner the best results.

Watch the complete video interview below, and be sure to check out more of SiliconANGLE’s and theCUBE’s coverage of IBM InterConnect 2017. (*Disclosure: SiliconANGLE Media’s theCUBE is a media partner at InterConnect. Neither IBM nor other conference sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE