The term cloud computing is bandied about all the time these days, but many folks are still confused about what all the fuss is about, and what it means to them. The IT landscape is changing faster than it ever has before, and it is becoming more and more difficult for quote “normal” people to keep up.
Unless you are an uber-geek – think Scotty from Star Trek -- and like to spend all your time reading through technical manuals, it can become real easy to fall behind in the skillsets needed to stay relevant in this new era of IT. As an example, storage area networks (SAN) are now considered traditional technology, and the SAN is being phased out as more and more solutions begin to adopt storage solutions which bring the data I/O closer to the system itself. Cloud infrastructure, HADOOP, GRID computing, and modular building block data centers are replacing the old Fibre Channel LUN (Logical Unit Number) based storage solutions.
New (actually old, but resurrected) technology like erasure coded disk is starting to impact the way shops are deploying RAID storage. Object-based storage and network attached storage storage is becoming more and more prevalent as application servers are virtualized into the cloud. Although good old SAN storage is still the best way to move large blocks of data around quickly, you still need to be aware and understand the changes in the storage and server landscape, and how it applies to how information technology is implemented today, and what it will look like in the near future. So let me try to explain what is going on as simply as I can. I will collapse all the elements that make up cloud computing into a single easy to remember term: intelligent abstraction (IA).
Cloud computing uses something called virtual abstraction to enable the rapid deployment of applications and data to reduce the cost and complexity of providing the underlying infrastructure, which also simplifies operations. The goal is to free up the IT team to get back to more strategic projects, and allow them to use technology as a service rather than something that they build and manage themselves.
My Definition of Intelligent Abstraction
I define intelligent abstraction as it pertains to the computer industry as the de-coupling of applications and data from all physical constraints while providing policy-based management, movement, storage, and protection of systems and information.
The concept of IA is simple. The cloud is made up of server and storage virtualization solutions which are tied together with software management and monitoring functions. According to the National Institute for Standards and Technology (NIST), for a solution to be considered cloud, it must include the following five essential characteristics:
IA combines the benefits of artificial intelligence (AI), server and storage virtualization, and policy-based automation together with advances in data management technology designed for the software defined data center to re-define how information and data center infrastructures are managed.
The great news for the storage guys is that since the underlying storage infrastructure is virtualized, it dramatically reduces the requirements for developing a deep knowledge of the storage itself. You only need to learn how the virtual abstraction layer works. The good news for the CFO is that since the storage is virtualized, it can come from any storage provider, which means you can pit your vendors against each other so you can purchase storage at the lowest possible cost. In essence, intelligent abstraction commoditizes storage and servers, and enables IT to become a service, which is the essence of cloud computing.