Mainframe and midrange server are probably among the least understood terms in IT – due in part to the fact that folks don’t necessarily know what differentiates these types of systems from each other.
Unless you work with mainframes or midrange servers specifically, you probably just think of both as weird, old-fashioned types of infrastructure. You know they’re different from commodity, x86-based servers and PCs, but you don’t know what really defines a midrange server or a mainframe, or what makes them different from each other.
If that sounds like you, keep reading. Below, we explain what mainframes and midrange servers are, and offer some examples of each type of system.
What is a mainframe?
A mainframe is a large (though not necessarily as large as you think) server that is designed for processing and storing massive amounts of data. Mainframes originated in the 1950’s and have been an important part of the computing landscape ever since.
Most mainframes today are manufactured by IBM (although that was not always the case). They run either z/OS, the native mainframe operating system, or a version of Linux designed for mainframes.
Despite common misperceptions about the decline of mainframes, mainframes continue to see widespread use across a range of industries.
What is a midrange server?
A midrange server, meanwhile, is a server whose processing power falls somewhere between that of a mainframe and that of a standard commodity server.
Midrange servers were originally intended to meet the needs of small and medium-sized businesses, which did not require the massive computing power of a mainframe but needed more power than commodity servers could supply. (Actually, for the first couple decades of their history, there was no such thing as a commodity server; x86 servers came into widespread use starting only in the later 1980’s.)
Although midrange servers operate using their own class of hardware, it is more helpful to define midrange servers in terms of the operating systems that run on them rather than the hardware that they are built with. IBM created the first midrange server operating, called System/3, in 1969, which gave rise to the concept of midrange servers.
Several new midrange operating systems followed until 1988, when IBM released its AS/400 server line and the OS/400 operating system for it. By the early-2000’s, IBM was callings its midrange servers System i, and the operating system for them IBM i.
You might hear IBM i (or iSeries, a closely related term) used today to refer to midrange server operating systems (you might even still hear people talk about AS/400). Technically, however, since 2008 IBM’s nomenclature for midrange servers has been Power Systems. That’s the more proper term to use today.
No matter what you call them, midrange servers continue to cater to smaller organizations that need significant computing power and availability, but don’t want to move their workloads to commodity servers or the cloud because of cost, performance or security concerns.
Midrange servers vs. mainframes
If you’ve read this far, you know that both midrange servers and mainframes offer higher availability and performance than commodity servers. But what makes them different from each other, apart from the fact that mainframes offer more performance?
Here’s a basic rundown of the differences between mainframes and midrange systems:
Mainframes have been around longer. As noted above, they originated in the late 1950’s, compared to the late 1960’s for midrange servers.
Mainframes and midrange servers use different operating systems. Mainframes run z/OS or Linux; midrange servers run IBM i.
From a hardware perspective, a mainframe is really a distributed network of components, which interact to form a massive computing platform. A midrange server is a single, standalone system.
Mainframes are usually larger in terms of physical size – although as noted above, modern mainframes are not the huge machines of yore; they can be as small as a refrigerator. But midrange servers are typically even smaller than that.
Despite these differences, mainframes and midrange servers both remain an important resource for businesses across the world. While the advent of the cloud and very inexpensive commodity servers has drawn some businesses to redeploy workloads to those environments, mainframes and midrange servers continue to offer unparalleled levels of availability and performance.
Read “Getting the Most Out of Your Mainframe” to learn how to offload, accelerate and lower cost while leaving the primary CPU with more headroom for the organization’s core business applications.
The post Mainframes vs Midrange Servers: What’s the Difference, Anyway? appeared first on Precisely.