A conversation with Thomas Taroni: Data Containers & Availability

Jim: “So Thomas, Phoenix Systems has already had a tremendous impact on the Swiss tech sector, and you obviously have aspirations beyond Switzerland. Given how quickly you’re expanding your operations, there are surely a lot of challenges that need to be met, data management being one of them.

What would you say are some of the biggest data storage challenges you’re facing today?

Thomas: “If we have a look back five years, you have machines with maybe 32 gigabytes of RAM by default. And if you have an unstructured database such as the one we are running, then you need a lot of memory. So one of the big challenges was to have the right machines in our data center to run these unstructured databases with the performance they need.”

Jim: “And what made you choose LinuxONE as your solution?”

Thomas: “Well, LinuxONE has helped us address several problems. One of our requirements is having enough power in place. The second is having the ability to automate our applications through containerization. That’s also a very important thing. And the third one is privacy. If you talk about containerization, you have to be sure that no one can access anyone else’s container without permission. So inside the LinuxONE, we have hardware designed to guarantee this. And that is very important for our customers.”

Jim: “So, on the subject of Phoenix’s big data requirements, can you talk a bit about Swissdox?”

Thomas: “Of course. Swissdox is one of the biggest solutions that Phoenix Systems is currently running. It’s an unstructured database for all of the newspapers here in Switzerland. We collect articles every day and store them in our archive where roughly 220,000 Swiss journalists and students can use them for research. It’s the largest media archive in Switzerland, containing articles going as far back as 1911, and collected from nearly 300 different media sources. Today, there are roughly 25 million different articles available, and another 1.7 million added every year.”

Jim: “And that obviously represents a massive amount of data. You’ve built an API that can search and serve up articles from an archive that size, but how do you plan for an evolving interface over time?”

Thomas: “With what we’ve created, we have a microservice architecture in our stack. So that means if you have a long-term project like saving articles, you have to be sure that these articles will still be available 10 years later, also running on the same system. The front ends are really changing over time. You know you’ll need to create, maybe every two years, a new front end. So you need an API where you could call all of these

requests so that in the end, you could change your strategy every two or three years as a company, and it’s no problem. It’s very important that you have your data in a long-time archive.”

Jim: “Given the amount of data you’re dealing with across your various lines of business, you have to ensure that these data systems are compartmentalized and secure.”

Thomas: “Yes, if we talk about our big data systems, in addition to Swissdox, we are currently employing a blockchain system for the Swiss government. Maybe the government is interested in these newspapers or might want to provide these newspapers to Swiss citizens. In that instance, we could connect these systems. Every business is of course concerned for its own privacy, but on the other hand, each one is potentially interested to have access to another. You build these big data systems, and in the end, they want to connect. Everything can be connected now.”

Next month, Thomas shares his perspective on the development and future of blockchain.

Want to learn more about how Phoenix Systems is changing the tech landscape in Switzerland and beyond? Check out Thomas’ profile in the Washington Post Brand Studio, or watch the video on how he’s using LinuxONE today.

LinuxONE is helping entrepreneurs like Thomas turn their visions into reality. Read the IDC paper about containers and Kubernetes, or learn more about the containerization capabilities of IBM LinuxONE.