The term cloud is being used in the marketing materials of just about every IT solution vendor in the world these days. Most CIOs and industry experts are falling over themselves to be part of the cloud before they get left behind. It’s to the point now that if you are still deciding whether or not this cloud thing is right for you, then you must not be that smart, or perhaps just have not kept up with technology.

The truth of the matter is cloud computing does have many advantages over traditional approaches to information technology, but it also has drawbacks.

According to the National Institute for Standards and Technology (NIST), in order for a solution to be considered a cloud solution, it must include five essential characteristics:

  • On-demand self-service
  • Broad network access
  • Resource pooling with location independence
  • Rapid elasticity
  • Measured service

If you look closely, you will realize that virtualization quickly becomes a requirement, and needs to be a key part of the offering.



Virtualization provides the capabilities for resource pooling and location independence. Think back to the late 1990s, when was just becoming huge, and vendors sprang up overnight to provide Storage as a Service (SaaS) solutions. The reason the SaaS vendors failed back then was based on the lack of the two fundamental requirements of network bandwidth and storage virtualization.

Back then, there was a lack of the high bandwidth optical network links to provide the basic services with the performance required by endusers, and their storage area networks required them to purchase like-for-like storage arrays, which matched their clients so they could move data back and forth when required.

Today, there is plenty of low latency bandwidth available, and now storage virtualization is used to provide a level of abstraction between the actual storage hardware and the services provided by the cloud providers. Virtual storage provides a common abstraction layer across heterogeneous storage arrays to enable the movement of the data from any client’s storage to the storage in the cloud. In some circumstances, the abstraction layer may also contain the intelligence to provide unified data continuity and protection functions across all the underlying physical storage, all without worrying about differences in the storage infrastructure being used.

Since the storage is virtual, it becomes much easier to manage, and it enables data to have complete mobility across any underlying storage hardware being used. The intelligent abstraction layer decouples the applications and data from all physical constraints while providing unified policy based management, movement, and protection of the stored information. The end result is a more cost effective, agile, and automated IT infrastructure delivering the required services to the business as efficiently as possible at the right cost. With intelligent abstraction software in place, storage becomes a commodity, just like server virtualization software makes servers a commodity. Intelligent abstraction is a feature of the software defined data center, and is at the very foundation of what makes cloud storage viable.



When server virtualization and storage virtualization are tied together with software policy-based management tools, the entire data center can be controlled and managed from a single console. This capability is now being called the software defined data center. Software-based commands from the management console enable the creation and movement of servers and storage pools on the fly, as they are needed.

A major benefit of the software defined data center as it relates to storage is the ability to include other data services as part of the resource pool being provided. As an example, let’s say you currently own your own data center, and have all the hardware and software you need to provide application services to your endusers. There is a cost to provide these services. First, you need to buy, install, tune, and manage all the servers and storage required for production applications to run. Next, you need to buy and manage the backup software and hardware, and all the capacity or media required to provide backup and data retention, including any data archive requirements which may be mandated by the government. Then you need to buy another data center, and install a mirror copy of your production environment in order to provide disaster recovery services for your applications, and lastly, you need to hire, train, and pay the staff needed to manage everything as the business grows. This is all on top of creating and running the strategic IT services and applications, which differentiate your business.

In some industries, this is fine, because IT is part of the actual products being provided to customers. Some examples would be Amazon, Netflix, UPS, and Federal Express. Amazon started as a way to buy books over the internet. Netflix began as a way to make it easier for people to watch movies vs. driving to the store to rent VHS tapes. UPS and Federal Express ship packages, and the ability to monitor and manage the shipping process is what they use to compete against each other, so IT is an extremely important part of their business.

Amazon became so good at providing IT services internally, they decided to provide their IT infrastructure to others as part of their cloud service offering. Netflix ended up becoming one of Amazon’s largest customers for their cloud services, and UPS and FedEx became the delivery arm of both. The genesis of the cloud was the idea of combining all of the best concepts and practices used by the experts who did IT for a living, and package it up into a service that could be purchased or leased by other companies.

After all, why buy a data center and fill it up with all your time and money, when you can just lease it from somebody else, and let them worry about it? The concept of moving IT to the cloud lets companies focus on their core business, and moves the expensive and time consuming task of providing IT services to the experts.

While you could build your own power plant to provide electricity to your house, you had better understand what you are doing, as there is a real risk you could get electrocuted or cause serious damage to your home. This same concept holds true for the other services you have and take for granted. When you pick up your phone to make a call, you should not have to worry about everything that needs to work perfectly every time you make a phone call. When you plug in a lamp, the electricity better be there to turn it on.

The concept of buying a service instead of building it and running it yourself is the underlying foundation of cloud computing and cloud storage services. It becomes much easier to move everything to the cloud, and let someone else worry about it for you. It also frees up the capital dollars used to purchase the networks, servers, storage, and other technologies required to make it all happen.

Once your data is in the cloud, other tasks such as data protection, data archiving, disaster recovery, business continuity, and technology refresh are removed from your plate. You just need to make sure that the contract you sign with your cloud service provider includes those capabilities, and the provider can offer that service to you at a lower cost than doing it yourself. The cloud becomes a no-brainer for smaller organizations who may not have the expert IT staff to begin with.

When you add it all up, the resulting savings from not having to buy, build, and manage IT can be huge!

If you are skeptical and hesitant to move everything to the cloud, you may still be able to get your feet wet by first moving to a hybrid cloud model, where you keep everything in your own data center, and ask a provider to just take over the tasks of data protection and disaster recovery. There may be ample cost justification to move just those tasks to a provider.



The cloud is not a panacea, and you need to be extremely careful about how you choose providers, and the details in the contracts you sign with them. With cloud, the contract is everything. Cloud storage services should specify stringent performance and latency parameters in the contract, and have financial or other remedies in case those requirements are not met. Good storage capacity management and capacity planning can become a significant requirement, especially if you have not done a good job in the past. Explosive data growth, which cannot be anticipated or accounted for may cause your costs to rise quickly.

You should have tools in place to ensure you can manage data effectively, and be able to delete or eliminate data which is no longer relevant to your business. Also, some cloud providers may charge you more to take your data out of the cloud than put it in. Be careful, and make sure you understand the contract and the services and costs associated with each. In some cases, it may be more cost effective to provide the service internally than using a cloud provider. You need to do the financial assessment work in order to become an educated consumer.



Will the cloud continue to rise, or will it disappear like what happened in the bust? Only time will tell, but the business world, and even entire governments, are adopting it quickly, as benefits of the cloud are just too great to ignore. Also, the fundamental technology requirement of intelligent abstraction is finally mature enough for it to last!