azure storsimple

Not long ago a fellow colleague, a CIO at his company, complained to me, saying: »Nowadays, nobody is deleting data«. The agonizing expression on his face was convincing, which gave me food for thought. The first thought that would usually come to me is something in the lines of “You don't need to delete, you simply store everything.” But all of a sudden, this didn’t feel like the right answer. I went on to explain to my colleague that the problem lies not in the amount of data but in the speed in which the data in being produced. I urged him to tackle this problem in a more resourceful manner.

A Cisco research determined that in the last two years we produced as much as 90% of all data present today. This sounds incredible. Different statistics, including mine, further determine that about 70% of user-created data remains in the same place where we stored them initially, even though we do not return to use them again.  Companies are often restrained by certain regulations and legislatures, limiting them from  deleting some types of data. And sometimes companies put such limitations on themselves, because they wish to process this data later on. Having said that, it is still wise to transfer data to cost-efficient storages. For example, into the cloud where storage is »practically free of charge«.

The problem of IT specialists and proprietors lies in the fact that they wish to store data nearby, sealed in a “box” under the table, preferably. Well, this will cost you quite a large sum. The trouble of managing data cannot merely be resolved by regular and/or periodic procurement of disk systems. Implementation of data storage systems isn't running at the same pace as data is produced in the most intensive and fast-producing data environments. Implementation is slowed down by the lengthy procurement process—first getting the CIO to decide on the system the company needs, then getting approval from the management and the Chief Financial Officer (CFO), preparing a tender, and selecting the right provider. Only then can the implementation begin ... and last for weeks to come. Time is more foe than friend to storage systems—a true curse. The rapid development of technology and frequent changes mean that the implemented solution will become outdated relatively soon, as often as after three years.

We must consider TCO also in data storage

Data produce cost. On all levels— acquiring data as well as storing and processing it. Based on previously mentioned facts, local storage of large-scale data is the most expensive option available. However, storing data is not the only thing that will cost you. For the most crucial data (at least) a company needs backup copies. One copy, as a minimum.  To all those who want to rest assured that their data is well-protected I recommend an even more thorough approach to this task – a company should make a production copy of data, a backup copy and an additional copy at a remote location. This can be cloud-based, since this is the least expensive choice.

Proprietors and CFOs are tough on Chief Information Officers, since they see data storage only as additional cost, but they often don't take into account the calculations of the total expenses of owning instead of renting storage space. If they examined this point-of-view, they would definitely move large amounts of data into the cloud, to other storage providers. They can offer extensively better prices of storage due to the sheer amount of data. At the same time, data is more secure being stored in their clouds than in the average company, since cloud providers also have their own experts as well as professional cloud solutions.

Why does the industry prefer renting storage? Mostly because companies are aware that the cost of data storage is not only about the procurement value of the equipment. Roughly speaking, this is only a third of the entire expense, but since this is the most tangible cost of them all, companies want to decrease the price of equipment as much as possible.

The future is hybrid

IT specialists and company employees will have to face each other and start up a dialogue. IT specialists are presented with the task of establishing what types of data they are dealing with and it is the employees' job to let them know how they want to process this data. As far as data storage security is concerned, the most important information employees they need to convey is how quickly they need their data restored in case of disaster or accidents, since the quicker the data needs to be restored, the greater the  cost of data storage. 

Computer capabilities and entire IT infrastructures are placing their bets on hybridity. Hybrid data storage is most often a combination of local and cloud-based data storage, but it has a variety of usage types, one of which is also local-only data storage, where companies use both faster (more expensive) and slower (less expensive but with greater capacity) data storage solutions. As far as price and operation adaptability is concerned, the most efficient choice of storage is definitely a hybrid solution, combining fast local storage for processing key data alongside a secure cloud data storage with an external (cloud) storage provider. This makes a company ready for all future challenges. The manager and the Chief Information Officer can now sleep peacefully, while the CFO is calculating the savings.