Cloud Computing and Virtualization are clearly a couple in the “Next Big Things” in technology these days. What I would like to do is to take up the topic of what it really means to virtualize information.
Just so we start off on the same page, here are some of MY definitions of the key technologies I’ll likely reference today and in the near future:
Virtualization - Isolation/separation/abstraction between in place resources and delivery of services from these resources
Infrastructure Clouds - An aggregation of resources that can be allocated and delivered dynamically (typically compute and storage resources)
SaaS - Delivery of an application over the network as an on-demand service in which the service can be leveraged across multiple separate “tenants” or clients
Information Virtualization – the ability to manage, protect, secure, and present information to multiple applications while also abstracting where the information is physically located
One problem that is hitting us with full force today is basic information management; mostly because the information (both content and data) is everywhere. It is buried in rows and columns in databases, hidden in proprietary formats and embedded within applications, and in file systems and repositories… everywhere.
If you look at the web today and how we manage information, we improved our ability to manage and find information by simply making copies of the data. We don’t move the data; we simply aggregate and index. This is what we value from the likes of Google and Yahoo today. We call it search but, really, the technology is just all about making copies.
So now it might appear as though we have a simple model to apply to all information, however this approach presents a few pretty big glitches. First, the format of the web is fairly consistent (using XML) so it is readable. Second, while there is lot of information on the web, it still dwarfs the amount of private information that exists. Even if we wanted to – there is not enough storage in the world to make more copies of everything. Third, unlike the web, this information is mostly private and needs to be secured.
Take, for example, the basic concept of creating a personal health record. One might think this would not be too hard; just get my health data, put it in a file, and put it somewhere – and voila – we are done. If it were only that simple… Think about it, our health information is everywhere (blood tests here, MRIs in that system, doctors’ notes somewhere, probably hand-written), and in most cases, it is not going to make sense to ever try to move it. This is where the concept of “Virtual Information” comes in, for transparent delivery and management of secure content and data, independent of any specific application. This is the real deal because organizations can leverage value of the content and data across multiple applications, and the cost savings through effective policy management can be enormous.
In future blogs segments I’ll share some of Information Virtualization capabilities we’re actively building across EMC. It is really exciting stuff!