Despite all the attention big data and Internet of Things (IoT) initiatives currently get, IT professionals still site users as one of the primary causes of unstructured data growth. Despite the fact machines and devices can create data non-stop, user data is more complicated to manage and less likely to be deleted. As a result, most IT professionals are less likely to remove old user data from their systems. Even archiving seems to be out of the question. The fear of managing user data is driving a sharp increase in NAS Sprawl.
Most organizations indicate machine generated data like sensor generated data, rich media (video, audio, surveillance) and log data is the biggest consumer of NAS capacity. But machine data is easier to manage. First, machines don’t complain, assuming good programming they will wait for the retrieval of data. Second, machines do a much better job of classifying data with tags that allow the management of large batches into a single operation. Third, there are typically better documented best practices on the length of retention for this data.
By comparison, user data is the wild west. Users tend to complain if they have to wait more than a few seconds longer than what they are used to. They tend to do a very poor job of organizing and classifying the data they create and how long to retain user generated data is often more of a best guess than a well documented practice. In response IT professionals feel compelled to keep old user data in the exact same place it was created, the primary NAS store. When that NAS reaches its capacity or performance limits IT has to buy another NAS system and NAS sprawl begins. Before the organization realizes it they are managing a dozen or more different systems.
In our webinar “End NAS Sprawl – Gain Control Over Unstructured Data“, we discussed how the varied unstructured use cases (big data, rich media, IoT data) leads organizations to buy NAS systems for each use case. But NAS sprawl driven by user data creates NAS sprawl within a single use case. Of course most organizations have those other initiatives, leading to an exponential NAS sprawl.
As we discuss in the webinar, IT can not accomplish the goal of eliminating NAS sprawl by throwing more hardware at the problem. Instead IT professionals need to look at the root cause, the file system. File systems are the core component of a NAS system and typically are isolated to a single piece of hardware or in the case of scale-out NAS, a single cluster of servers. In the webinar we discuss how file systems need to evolve, shed their hardware dependency and work across disparate hardware and even locations in order to create a more organized workflow.