We already often talked about cloud computing on this blog and about the transient nature of executing programs in the cloud. I also mentioned that data is something completely different, in the sense that data in the cloud is persistent in nature. I believe many cloud users do not have a complete view of the implications of putting their data in the cloud. For this reason I developed a point of view on the subject, titled "Data and the Cloud" that I append to this post. Feel free to download it. I may not have addressed all aspects of the question, but believe the document contains a fair amount of the issues that need to be addressed.
From a cloud perspective, there is no need to understand where the datacenters powering the cloud are located, but from a legal perspective, one better knows where all copies of the data are stored. In particular when in concerns privacy, legislation is there to specify what can be done and what not. You may argue that these laws are old fashion and should be adapted to address the needs of today's global world, and you may be right. Unfortunately, countries have different views of their citizens' privacy rights, making the globalization of the law difficult.
Legal issues are not the only ones that should be raised when speaking about data in the cloud. There is also the whole security aspect. In a blog entry titled Cloud Computing - "Data in the Cloud" podcast with Dana Gardner ..., Archie Reed, from our security office, addresses some of those aspects in quite some details. Illegal data access & data proliferation, are elements to take into account when moving data to the cloud.
The cloud model allows you, as Charles Babcock points out in his new book, "Management Strategies for the Cloud Revolution", that with simple devices you are now able to interact and manage extremely large compute capacities in a peer-to-peer capacity. And this is where he puts the cloud revolution. That is perfectly correct, and the book is definitely worth reading (which I am in the middle of doing), but I would like to add one element. By going back to a model where the computer power and information is centralized, we host our knowledge in a limited amount of places on the globe. This may make us more vulnerable to attacks from all sorts. As cloud service providers are extremely secret about the precautions they take, we have to trust them.
This would not be an issue, if at the same time we were not in an increasing digital world. Photos, films, music, literature, art, everything is digitized. As Tom Freedman points out in his well known book "The world is flat", everything becomes global, digital, mobile, virtual and personal. So, we increase the digital nature of our knowledge, our history, our world, and store it in a small amount of locations around the globe. That's one of the reasons why data in the cloud has to be looked at so seriously. These datacenters are actually doomed to explode in size, knowing that the digital information doubles every 18 to 24 months.
Enterprises want to make sure the information they need to run their business is kept securely, is not accessible by their competitors, is destroyed completely when requested and is not proliferated around the cloud. That obviously makes sense, but as we saw before, I believe that this is only one aspect of the issue. The more I read on the subject, the more it becomes clear to me that the issue is way larger. There are obviously a lot of positive elements of maintaining data in the cloud, one of which is the ease by which it can be accessed (by entitled people) all around the globe, facilitating collaboration and people interactions. Let me know what you think.
