"I think there is a world market for maybe five computers," goes the famous quote attributed to IBM Chairman Thomas Watson.
IBM's most advanced computer bears his name, so it does seem like a crazy prediction.
It is not true. Not only because there is no evidence he ever said it, but because the prediction may have overstated the need by four computers.
Cloud in computing terms got its name from the icon used by IT designers to draw the elements of their networks that connected to the internet or that they did not manage. It stuck, even though many non-technical types might think it might actually have something to do with real clouds.
Cloud storage and cloud computing
Cloud storage uses the more secure, vastly greater size and typically lower cost of storing anything on a remote storage device rather than on your local computer or phone.
If you have a Dropbox account, welcome to cloud storage.
You may not realise it but if you have an Android phone, you probably store your photos on Google servers.
If you are a Windows user, you have a cloud-based storage option called OneDrive, even if you did not realise it.
The principal value in keeping, at least, a copy of your important data in the cloud is two fold.
If you lose or damage your local storage, you will still have your copy.
- Businesses are possibly more at risk with only local storage as ransomware targets businesses with malicious software that locks access, demanding payment to unlock it.
You data is also less likely to be stolen when using a secure cloud option and if you have a lot of data to store then it will probably be cheaper too.
The downside though is the bandwidth cost for moving data to and from the internet.
Cloud computing is the more significant of the two. The adoption and use is growing.
When you visit your Facebook profile you are partly simply accessing a website, but for videos you edit or albums of photos you enhance, you are using Facebook’s computer's processing power.
If you use Gmail or Google’s suite of document applications, they are both examples of cloud computing.
Windows too is moving its Office products from something that is run on your machine to being run on their servers.
The largest platform is Amazon but this would be a business solution, even though some of your favourite services run on them.
The principal advantage is that their products will never appear underpowered or slow as they control the power of the machines it runs on. It also allows them to fully control access to the service which is a useful business strategy .
Software is considered a product but the growth of the cloud makes more software, in effect, a service.
Anything could be moved to the cloud with the most intense processing options for things like games being perfect for cloud computing provided that the internet connection is fast enough with low latency (the time it takes to respond).
An indication of the demand for cloud options can be seen in the annual LinkedIn most sought after skills, which once again has cloud and distributed computing as the most in-demand skill.
The Internet of Things (IoT) is also highly reliant on cloud storage and computing by keeping the devices “dumb” which lowers the cost while maintaining their function.
With so many dumb devices connecting to a massive cloud based distributed computing unit, you can see why the global market for computers might actually just be one, a big one!
The LinkedIn Top Skills of 2016
But there is one very large issue to be resolved before we can all head to the cloud. Your ability to do anything is determined by your access to the internet and should that be cut, you are lost.
Actually, cutting cables is not easy as internet traffic has multiple routes across the network.
The real threat is a Distributed Denial of Service attack (DDoS).
Recently an attack knocked out access to some of the world’s most popular websites and services.
The Australian government was embarrassed when their census poll crashed following a DDoS attack.
The attack occurs when a web server has more requests to return a response than it is able to.
Assume you are talking to a shop assistant, you make your request and they respond with the information or product you want. If there is a queue, each is dealt with in turn. If everyone in the queue demands to be served at the same time, the assistant will not be able to respond to anyone.
Typically, creating enough traffic to overwhelm a server is not easy; they are designed to account for spikes in normal internet traffic. The attacks require huge increases in traffic.
The source of that traffic in the most recent attacks are the “dumb” devices that have been hacked and instructed to act in unison to access a specific site at the same time and to do so repeatedly.
It illustrates just how many of the devices are already on the network.
Remote cameras and routers make up the bulk of the devices, but anything with a weak password and the ability to connect to the internet is a potential host forming a network of remotely operated devices called botnets.
The issue is a significant one.
Google has built a system for its own protection that it extends to media and organisations that are at risk of being censored.
The best way at the moment to deal with attacks is to both have sufficient ability to handle traffic spikes and to determine if the traffic is a real request or part of an attack. The system can reject traffic that is deemed suspicious.
It is an easier way to deter DDos attacks than the attempt to update the millions of aging internet devices that are being hacked.
Future “dumb” devices will have better security to further lower the risk, at least, until the next threat is created.
Threats aside, the advantages offered by cloud services will see everything shift in that direction as fast as the internet access and speed will allow.
There will be a need for more than one computer, but would you really have read this far if the headline was, "multiple cloud-based computing options to determine the future of the internet"?