What role will data management play in a future which is increasingly being defined by the mass adoption of cloud-based technology solutions?
The fact of the matter is that the world is running out of storage space. Research undertaken by the IDC noted that capacity shipment to the cloud increased by 18.3% to 52.4 Exabytes in the fourth quarter of 2016 alone.
Further research by research firm Marketsandmarkets predicted that cloud storage will grow by a compounded annual growth rate of 25.8% from 2016 to 2021. In the end, this will become a $74.94 billion market.
And yet, data is becoming an increasingly important commodity. Insurers are after data to profile the way clients live in order to assess the risk that they are bringing on, and retailers are desperate for data on the spending habit of clients so that they can market specific products to specific groups.
This means that there is a specific need for data management so that all parties can benefit from this key commodity.
Restarting Data Management
So how would one readdress, or even redefine, data management? I read a recent article on enterpriseinnovation.net which provided some key insights into the subject.
The article pointed out that, in the past, most storage administrators looked to control data volumes. Different storage architectures were used to rein in storage growth.
The article adds that as today’s companies embrace digital transformation and increasingly rely on data to be efficient, data volumes are growing exponentially faster than ever before. A big contributor to this unrelenting growth in unstructured data, such as emails, videos, documents, etc. Many previous storage infrastructures were not specifically designed to handle this data type efficiently.
But managing data volumes is only one aspect of data management. Companies need to also handle variety and velocity of data as they shift to a digital-centric and mobile-driven environment.
Speaking to enterpriseinnovation.net, Matthew Johnston, Area Vice President, ASEAN & Korea at Commvault said that data management is a broad term. “However, what we are really talking about is the value of data. The reason we manage data is because it has value. And to understand the value you need to categorize it,” said Johnston.
See the big picture
The article adds that the cloud can offer much needed relief. But without a strong data management strategy, it is not going to solve all data management problems. Instead, Johnston believes firms should take a step back and start examining the data itself.
“First step is to understand your data intimately and its relevance to the business. All data are not created equal. Some data is more critical. So, you need to understand what can be placed in the cloud,” Johnston told enterpriseinnovation.net.
The article adds that a clear understanding of the data and its value to the company can help firms to understand where and how to store it. It also allows the firms to look at new ways to mitigate data-related risks in case the unthinkable occurs—and it does.
Consider the latest Amazon Simple Storage Service (S3) issue. A simple debug of a billing system took down the servers. The issue disrupted websites and businesses. If you are an e-commerce company, it may mean millions in lost revenues.
The article adds that, meanwhile, Gitlab, a well-funded and known startup that saw itself as an alternative to GitHub, recently suffered a similar outage when an employee deleted the wrong files. Although the mistake was unintentional, the damage was done. What was worse is that they had to inform their customers that data created within the past six hours were lost forever.
“All are offering is the infrastructure. The responsibility to look after the data is yours,” Johnston told enterpriseinnovation.net, adding that a sound data management strategy should be a firm-wide concern.
Johnston highlighted four key steps that firms need to consider when architecting their data management strategy: manage data by region; clearly know where your data lives; make sure you have an alternative data recovery plan; and have a holistic data strategy.
“Once companies understand the data and have the , it is then about choosing the right technology that gives them the flexibility to move the data,” said Johnston.
SDN to the rescue
Johnston urged companies to stop looking at data as a storage problem. He noted that companies need to move away from this short-sighted view when building their data management strategy—especially when considering the complexities of operating in a digital world.
“Take IoT for example. Knowing the location of your devices is only information at a certain point in time. To make better sense of the data, you need to store and correlate it, which can mean a huge volume over a period time. Also, you need to understand how you can protect and store it for longer periods to draw long-term correlations,” said Johnston.
The article added that he advised companies to take a hard look at Software-Defined Data Services (SDDS). It essentially abstracts the data away from the storage array, allowing complete portability of data across platforms whilst maintaining the same user and application experience. This expands the value of the data. It also means companies can create a virtual data lake without having to move all the data into a single repository.
“From our perspective, we become essentially agnostic. We support data regardless of where they are stored and in which format. It also does not change the policies on how you manage your data. What it gives you is agility from a storage perspective,” said Johnston.
Opportunities arise
While there is a growing need for data management in the industry, not every company has the skills to do it.
If you can position yourself as a company that can effectively collect and manage data in an effective manner, there will be ample opportunities for growth in the market. However, this needs to be done bearing the impending implementation of the Protection of Private Information Act in line.
Equifax, and most recently a cyber breach in South Africa, saw the personal information of citizens leaked onto the internet. This included information such as social security and identification numbers as well as bank account details.
Data needs to be separated according to importance, relevancy and sensitivity. Once this has been sorted, the data needs to be stored in a manner where the company feels that it will be safe. Trust is a major issue when it comes to data management. Companies need to trust that their vendors have their best interest in mind when it comes to data management. Once trust is lost, it is a hard task to reclaim it.