The impact of the Megaupload shutdown on cloud computing in business

The recent shutdown of file-sharing website Megaupload has created a sense of unease surrounding cloud-based storage. Companies are questioning the wisdom of relying on these services, which could seemingly collapse at a moment’s notice. The timing is unfortunate, given the advances the cloud is about to go through.

The recent emergence of cloud computing has greatly stimulated and facilitated resource sharing - beyond organisational boundaries - among various users. This resource sharing is primarily driven by resource virtualisation and utility computing (the pay-as-you-go pricing model).

The legal proceedings against file-sharing website Megaupload should not be seen as an indictment against cloud computing. This would be similar to discrediting the internet because a major ISP has gone out of business!

In certain businesses and industries today, clouds are practically taken for granted. Most start-ups today use available public clouds such as Amazon to drive their business. As time goes by we will start to see fewer companies insisting on owning their own server farms. Many venture capital firms today have cloud-based ventures in their portfolios. In the corporate IT world, what will drive the use of clouds is the wave of ‘big data’. Also, the emergence of private clouds is quite popular with corporate IT, especially developments related to Hadoop for distributed storage and distributed computing. These developments could have an impact on the use of private as well as public clouds.

Another promising approach for using clouds is what is popularly known today as cloud bursting’, in which private and public clouds are used to build hybrid environments. In this case a business can expand its storage or computation capability by using a public cloud during peak operating times or even during regular operations. In this approach, private clouds can elastically expand into a public cloud in a seamless way (and also shrink back into the private cloud).

The above is leading to a surge in storage R&D and we are experiencing a new rise in popularity (more than at any other time). A lot of funding is being injected into setting up storage start-ups and this is expected to rise. One of the motivating factors for this is the fact that moving massive data remains a major bottleneck and the best available approach is to move computation to where the data reside, rather than the other way around. I believe this would be a great push for more cloud solutions. For example, we can move machine learning algorithms around in a cloud platform to process existing data and generate more data that can be used to make decisions. Of course, this gives rise to many challenging issues that need to be addressed to enable this interweaving of computation and massive data to happen in an efficient way.

To go back to the earlier concern about the Megaupload case, I still feel this is a minor risk compared to what promises cloud technology hold. For example, one can use multiple cloud providers to mitigate the risk of one site going down or going out of business. Also, to use a mix of private and public clouds would be another approach.

To date, we have seen a lot of investment and innovation going into improving cloud offerings and I think this is bound to continue for the next few years. As a result, we should see more stabilisation of cloud technology leading to the emergence of more robust and reliable platforms. In a nutshell, clouds are here to stay, and they are not a here-today, gone-tomorrow phenomenon!

Albert Y Zomaya is the Chair Professor of High Performance Computing & Networking in the School of Information Technologies, Sydney University. Professor Zomaya is the author/co-author of seven books and more than 400 papers, and the editor of nine books and 11 conference proceedings. He is the Editor in Chief of the IEEE Transactions on Computers and serves as an associate editor for 19 leading journals.