Ohn Naughton in his blog article- 'Every Cloud Has a Silver Lining' says how data processing is similar to electricity- from being decentralised to the one that is centralised involving number of big companies. However, unlike electric current the data flows are unregulated and the big service providers are yet to be treated as utilities.The article in itself is an extension of Nicholas Carr's attempt to explain Cloud Computing to the masses entitled 'The Big Switch: Rewiring the World from Edison to Google'. Carr compares cloud computing to that of building an electric grid. Electricity was once local and every factory had a generator and then it was expanded and distributed through a national network known as the grid. Similarly, data was once local existing in the server rooms of individual organisations and now it is centralised and distributed in the "server farms" of international network (internet). Most of the services that we regularly use such as Facebook, Google, Instagram, Spotify, Google Translate, etc are possible through the computational powers of these server farms and the internet bandwidth. Our individual computers or computing devices are just like data switches that receives this flow of data when switched "on" to those services.
Storing our data such as audio, video, photos, documents have become all the more convenient as we need not rely on our external drives and in-built hardware of our computers. This is all due to cloud computing where information is obtained at one's finger tips. However, the flip side to it is that the security and the risk involved could be heavy as it is not under our control. If services providers encrypt data, it could also decrypt data. Plus they are vulnerable to attack to the extent that our data might get stolen or we get locked out of our accounts. Keeping this in mind, we have to choose wisely on what kind of data we wish to store in a cloud and what we are better off not doing so. Also, once we decide on data
storage, we need to research what kind of service is being provided and at what cost. Every service provider have their own costs for the kind of data that need to be stored.
Here are some key features of data management using cloud:
1) Know Your Service Provider: Features such as WeTransfer and Sendspace are really useful for sending big multimedia information from one party to another. This can be done for free. There is a membership option (premier membership) through which you are given bigger bandwidth to transfer data. However, for normal usage free service should satisfy your basic needs.
2) Reliability of the Service Providers. There are many players on the cloud providing different levels of support and different prices. However, the main players that would stand the test of time always tend to be Google, Facebook, Apple, Amazon and Microsoft.
3) Bandwidth Limitations: Bandwidth determines how much data can be transferred and the duration it takes to transfer. Hence bandwidth range, time and cost involved are important considerations.
4) Encrypt Your Data Yourself: A strong encryption of valuable data is recommended from the user's side instead of getting it done at the provider's end.
5) Read the Fine Print before Signing up: This can save you from signing up to something that is highly expensive, that takes no responsibility for data or time loss and where you compromise on security.
6) Read the Features offered by the Cloud Companies: Find out the bandwidth being offered, the data restoration and recovery time and the cost.
7) Having a Backup Plan: To switch from one cloud computing company to another with minimal downtime should there be a situation when you need to change business or the existing cloud provider closes down.
8) Effective Data Recovery and Restoration: Data restoration along with data recovery should be fast, efficient and manageable. Data backup and restoration can be easy. However, recovering it could be problematic. There should be an assurance given in terms of data recovery time or if they provide a hard drive along with the data.
9) Buy Only Storage You Need: Sometimes it is hard to make out the storage that one would really need. When talking of big businesses with huge volume of data, the storage capacity might be less important than the storage consumption. The company is very much likely to predict the annual pricing and number of users. However, it is not so with storage size. According to Hubert Yoshida, CTO of Hitachi Data Systems, buying with a pay-as-you go model could be more effective as data centres decline in prices every year.