Online backup is, to be honest, a rather nebulous concept for most of us. Sure it works, but what does ‘backing up to the cloud’ really mean?
One way to grasp the idea is to understand the history of backing up, because by understanding the evolution of the technology behind backing up data, it is easier to understand the current state of affairs. We therefore present a brief history of how the modern world of online backups came to be.
The Magnetic Tape
Magnetic tape was first patented in 1928 by German engineer Fritz Pfleumer (based Vlademar Poulsen’s invention of magnetic wire), but only entered widespread use as a medium for mass storage of computer information in the 1950s, where it played a key part in the ‘computer revolution’.
One of the first practical high speed tape systems was the IBM 726, which was announced on 21 May 1952, and which used a new ‘vacuum channel’ system to allow the tape to stop and start almost instantly. More importantly, it could store two million digits, which was considered a huge amount at the time.
The IBM 726 was sold together with IBM’s first digital computer, the Model 701, and could be rented for $850.
IBM introduced the first hard drive, the 350 RAMAC disk storage unit, in 1956, and by the early 1960s the hard drive disk (HDD) was the dominant backup technology, as it still is today. Hard disks store data on rotating rigid disks (known as platters) coated with magnetic material, and use magnetic heads arranged on a moving actuator arm to read and write data to the surfaces.
The IBM 350 was the size of two refrigerators and stored 5-6 million characters (in modern terms 3.75 MB) on stacks of 50 disks spinning at 1200 RPM, while modern hard disks usually come in a 2.5” or 3.5” or 5.25” form factor , typically store up to 3TB of data (1TB = 104,8576 MB), and spin at up to 7,200 RPM.
Seagate introduced the first hard disk designed for microcomputers, the 5.25” ST505, in 1980. It had a capacity of 5MB (five times greater than a floppy), and cost a whopping US $1,500 ($4,293 in today’s money). In the same year IBM announced its most successful mainframe hard disk, the IBM 3380, which became one of IBM’s biggest sellers of all time with over 100,000 units sold, and which could store a whole 2.5GB of data (later models could store up to 20GB), with prices for the base unit ranging from $81,000 to $140,000.
HDDs still dominate the mass storage market thanks to their high capacity and low price per unit of storage, and are expected to continue doing so for the foreseeable future As such, they form the backbone of modern cloud storage solutions.
As computers became more mainstream and available to a wider audience (although still at this time prohibitively expensive for home use), the need for more portable storage became apparent, and this need was met in 1972 when a team led by David Noble invented the 1 MB 8” floppy diskette.
Although originally designed for the IBM 3330 (‘Merlin’), the fact that this lightweight magnetic disk was cheap and could be easily transferred meant it quickly gained mass popularity, and became the default medium for computer storage.
By 1978 the 5 ¼” floppy disk (first introduced by Shugart Associates in 1976) had replaced the 8” disk in popular use, as it was considered a more suitable size for the new breed of desktop computers, and in 1981 Sony started to ship the 3 ½” floppy drives and diskettes (with rigid plastic jackets) that were to dominate the market for the rest of the technology’s life.
Still popular well into the early 2000s, the floppy disk became less of a mass storage device and more of a ubiquitous way of transferring personal files.
The idea of using light to record and replay music was invented in 1965 by James T. Russel, but it wasn’t until the late 1970s and early 80s that Sony and Philips began serious work on the concept. In 1982 the first music CD hit the shops, and in 1984 the first CD-ROM (Compact Disk Read-Only Memory) was produced, using same form factor as the music CD, but capable of storing computer data on tiny pits etched into the plastic of the disk.
It was not until 1990 however that the CD-Recordable (CD-R), also developed by Sony and Philips, was introduced, and the heyday of optical backup could begin. Cheap, easily transferable, readily disposable, and holding over 600 times as much data as a floppy disk (650MB), CD-R rapidly became the back-up solution of choice for home users.
Since CD-R’s introduction, increasingly fast write-capable drives, plus the introduction of new high-capacity optical media standards (such as DVD-R and Blu-ray Disc recordable (BR-R)) have sought to maintain the position of optical backup, but over the last few years falling hard disk prices (including the introduction of easily portable external hard drives).
Plus the introduction of solid state media (most notably increasingly cheap and capacious USB ‘thumb drives’), has started to see optical storage falling out of favor as the home backup system of choice. Most importantly, the rise of cloud computing has revolutionized the field of backup storage.
Cloud (or distributed) computing has its origins as far back as the 1960s, when large mainframes where accessed using ‘dumb terminals’, but it is only now that large numbers of people have fast broadband access to the Internet that the concept has really taken off and achieved mass appeal.
Online storage is a form cloud computing that concentrates on, of course, storage. When a file is uploaded to ‘the cloud’, it is transferred (usually in encrypted form of some kind) over the Internet to another computer (a server), which is attached to banks of hard disks.
The file is then saved on these disks, which are usually arranged in RAID arrays to improve performance, and mirror data across multiple drives to ensure against data loss. Large facilities housing many online backup servers attached to hundreds of hard disks are known as server farms.
As this history of backing up should show, none of the technologies used for online backup are particularly new, but it is rather the combination of faster and more generally available Internet, plus cheap mass storage that makes the concept both possible and compelling.
Rather than individuals’ backing up their data manually and storing it locally, for a few dollars a month the whole process can be automated, and vast amounts of data stored off-site (following the golden 3-2-1 principle a local backup should also be made, but this too can be automated, often using the same software that performs the online backups). We hope this history of backing up has been interesting, and will help readers understand how the world of online backup has got to where it is.