March 12, 2014 | Written by: Sujith Pillai
Share this post:
I have been traveling across various parts of the world to talk about cloud computing. I also post various updates about cloud computing on social network sites. This made my non-techie friends and family members ask me: “What is this cloud computing that you always talk about?”
Cloud can be compared with any industry that has changed from discrete business to social business. I’ll give you an example. In ancient India, Goshala (cattle farms) could be found everywhere. Everyday needs were fulfilled locally at every house, such as milk (for drinking, making sweets) and manure. The milk or manure that you get in excess is waste, as there is a limit to what you can preserve. If you compare this with the olden days of computing, the way of operation is exactly the same. Every firm had computers for their local needs. They didn’t care whether the energy spent to maintain the computer was being utilized in an effective manner. Only around 10-15 percent of computing power was used, and the rest went to waste.
An immense amount of energy is used to maintain a cattle farm. You have to feed, clean, nurse and milk each cow every day. This is similar to maintaining computers. You have to provide uninterrupted power, air conditioning, maintenance and security for the computers to operate well. These things are needed even if you have only one computer or thousands of computers. The cost and energy needed for maintenance is not proportional.
In modern times, people moved away from maintaining cattle at home. A few people started bigger farms to shelter cows. They started selling milk and other dairy products in bulk. The public could pay money and get the necessary things when they wanted and only as much as they wanted! Similarly, firms which had many computers have to pay the same amount to maintain the computers even if they are utilizing only 15 percent of the computing power. So they started sharing the leftover 85 percent computing power with others. They don’t lose anything; they gain money by selling the computing power. On the other hand, the buyers don’t have the burden of maintaining the computing system (or maintaining a cattle farm).
Let me clarify some characteristics of cloud computing with this comparison:
• Elastically scalable
Today, if you want to make tea for yourself or for 15 guests, you only need to go to a grocery shop and buy either enough items to make tea for one or enough for 15 people. Similarly, based on the demand, the computing power is available for you in the quantity you need.
• Pay per use and pay as you go (PAYG)
If you buy milk, you pay only for what you are buying. Similarly, you pay only for the computing power that you use.
• Self-service model and automation
You can buy milk directly. There is no need to milk the cow and get the milk pasteurized. Similarly, you choose what kind of computer you want by yourself. The cloud will automatically provision it for you with little to no human interaction.
Not only you, but others also can buy milk from same grocery shop! Similarly, not only you, but others also can get computing resources from the same vendor.
All milk that you buy from a shop has certain standards. You know the percentage of fat in it and the expiration date. Similarly, every computer you get from cloud will have certain amounts of processing power, as well as a certain platform and operating features.
• No capital expenditure (CAPEX) and variable operational expenditure (OPEX)
You don’t pay up front for the cow or the infrastructure needed to shelter the cow. You pay only the amount per gallon of milk. Similarly, you don’t pay the data center for maintaining the computers. You just pay for the computing power that you use.
In my next blog post, I will explain more about cloud delivery and deployment models with similar examples. What do you think is the best industry to compare with cloud computing? Do you think any best practices from these industries can be brought to cloud computing? Write your thoughts as a comment to this post or catch me on Twitter @sujpilla.