Few devices bring to mind something quite as immobile as a classic mainframe. Moving one is nobody's idea of fun. But join mainframe technology with the needs of mobile, and suddenly we have something that moves nimbly and with great power. Those who do reap serious benefits. Let's look at a few of these benefits, as well as a few real-world examples.
Mobile services must be able to handle dramatic variation in usage demands, with ironclad reliability. Sure, the cloud can scale, but each of the hypothetical 100 connected servers has its own potential performance issues to be fixed, let alone maintained. A mainframe is able to scale up or down instantly without the complications that come with relying on 100 different devices to provide the same performance.
Secure host required
Mainframes are famous for security. Some cloud service providers that offer hosting via distributed servers host their own data on a mainframe. You don't hear of many data breaches occurring on a mainframe. Mobile banking and e-commerce conducted through smartphones are among the most vulnerable types of computing transactions. To cite one type of consumer-based scenario, a mainframe-based hosting platform allows a hosting company, a merchant who hosts their site with that company, and the merchant's customers to share the same level of security and reliability.
Does security get more high-stakes than in banking? Italy's Banca Carige not only stores customer data (mobile and otherwise) on their mainframe for the highest level of security, but can also leverage the mainframe's processing power to conduct real-time analysis of customer data. The insights enhance security and provide the company with targeted opportunities to increase customer satisfaction.
Secure host required
I've already mentioned this remarkable advantage that mainframes bring to the mobile playing field: Mainframes are perfectly aligned with simultaneously running an app and securely serving the data it uses. This combination yields such opportunity for mobile-based businesses that it deserves its own discussion. The notion that data must be separated from the application using it is so common that people don't realize that things don't have to be this way. Why put layers between data and the app? For that matter, doesn't it make great sense to locate analytics software as close to its data as possible? Real-time data analysis needs every sliver of efficiency it can get. Mainframes make co-location of software and data simple, and provide many opportunities for greater efficiency.
Here's an example: First National Bank (FNB) is a leading bank in South Africa. One of the reasons for its success is its first-to-market mobile banking app. The app caught on quickly in no small part due to its speed and reliability. FNB's developers were already familiar with the ability of mainframes to scale, since the company had been using them for decades. The app's front end directly connects with the mainframes where customer data resides, eliminating data replication. All of these factors result in a mobile success story for FNB.
Anywhere, all the time
Mobile computing sets many expectations. Devices need to be able to access data from anywhere in the vast universe of servers and networks. Data everywhere must also be available all the time, in no time. No hiccups or milliseconds of waiting are acceptable to mobile users these days (admit it, you get impatient when a data call takes long enough for you to notice.) Most IT managers adequately manage to keep up with the demands of mobile usage, but wouldn't it make things easier if there was a solution with a decades-long track record of satisfying all of these requirements?
Mobile computing, meet the mainframe. You are a match made in heaven. Mainframes commonly have uptime rates of 99.999%. When companies that rely on mobile for their core business use mainframes, the result is something like that famous candy bar commercial where chocolate and peanut butter accidentally mix and create a powerfully tasty mix. In this case, the mixing of mobile and mainframes seem made for each other.
This article is published as part of the IDG Contributor Network. Want to Join?