Mainframes and Cloud Computing: Similarities and Differences

Since the arrival of cloud computing, the debate has raged: is cloud computing really just the mainframe re-imagined? Or are the differences significant enough that these two technologies should be considered opposite? In this article, we'll take a look at both the similarities and differences of mainframe and cloud computing technology, including scalability, security, and terminal emulation considerations.

What are the similarities?

Client-server model: If you look closely at both mainframes and cloud computing, you'll see that both are implementations of the client-server model where applications are stored and run on remote servers.

With the mainframe, applications are stored and run on large server computers stored in specialised, air-conditioned rooms designed specifically for housing the mainframe. Users then connect to the mainframe with the use of client terminals or, in more recent times, terminal emulation software. With cloud computing applications may be stored on various servers, but they are still accessed through a thin client).

Thin clients: Because all of the data is accessed from a remote location (ie. it is not stored on the user's own computer), users are able to get faster and more painless access to data from wherever they are located. Admittedly, these similarities are based on the differences in the computing environment in which they emerged.

The mainframe, for example, was centralised because, several decades ago, computers needed to be big enough to fill entire rooms in order to harness the required power. "Dumb" terminals were used to access the mainframes, which is why today terminal emulation software such as 3270 emulation is required to access the mainframe. However, in the modern computing environment, where laptops, smartphones and tablet PCs are becoming increasingly dominant as a means of accessing servers, data now needs to be stored in remote locations for storage and security reasons.

What are the differences?

Server location: When it comes to a mainframe, the server is always a single, monolithic host that resides in a centralized location -- as mentioned earlier, this is usually a specially-built and air conditioned room to house the mainframe. Cloud computing, on the other hand, may be scattered across many individual servers.

Scalability: One of the main advantages of cloud computing over mainframe models is the ability to dynamically scale resources and power. Because the cloud can be accessed from anywhere, businesses can increase access to the server as they grow simply by adding further clients. They can even extend this access to mobile devices.

Security: One of the biggest advantages of mainframe computers is the security. Because cloud computing relies on an internet connection in order for users to access the server, it runs the risk of being compromised by intruders. Additionally, by utilising a cloud service provider, you're effectively putting your trust in that service provider that your data will be safe, as it will be kept on servers that are owned by them.

Reliability: Another advantage of mainframe computing is higher reliability. If an internet connection goes down, or if the host servers go down, the client will be unable to access its data. At least if there is a problem with a local mainframe, there will be on-site IT staff available that can be working on the problem as a priority.