Today, traffic cameras, body cams, motion sensors and more are everywhere. Internet of Things (IoT) devices are everywhere, generating more and more data. According to a recent report by IDC, the world will create 44 zettabytes of data by 2020, and 80 percent of that will be unstructured. Beyond the high amounts of data traveling over networks, another growing struggle is finding where to store all this data (and how to back it up).
While some newer, web-native companies use object storage to manage their content and move information between data centers or multiple clouds, most established enterprises are stuck managing many disparate groups of data—some even still on tape. Enterprises feel the drawbacks with tape, which aren’t always reliable and can cost precious time. Each tape backup requires someone to look up the serial number, track down the tape, then physically retrieve it to access the data.
Why object storage instead of file or block storage?
In the past, file and block storage (or NAS and SAN) made sense when on-premises storage was dependent on locality for performance, and growth was more gradual. However, we now live in an age when no one deletes anything. Especially as teams become more global, large files and projects require unwavering accessibility and speed across distributed access points. File storage’s hierarchy tree can slow down when it needs to manage billions of files, and both file and block storage can’t scale as well as object storage. Compared to file and block storage hierarchies, object storage is flat, making it simple for IT teams to access individually-tagged files quickly.
Compared to other types of storage, the cost savings from object storage is very predictable. Primary storage costs net out to the overall cost of data centers as well as the spend needed to manage them. Rather than worrying about the extra maintenance and management for various structures, organizations can merely pay as they go with object storage.
Where does IoT stand to benefit?
As more internet-connected devices come onto the scene, enterprises will want to develop applications for them and analyze the data they produce. Object storage makes it easy to centralize IoT application development. The two also play well together by design; since object storage was developed for the internet, it’s extremely HTTP-friendly—just as IoT devices are. Because both object-based approaches and IoT devices are based on RESTful APIs, it makes it easier to push and pull information, which can be compelling for the size and speed needed for data analysis.
Who’s using object storage?
Law enforcement is an industry that’s collecting huge amounts of body cam video to catalog and manage where they’ve never needed to before. Another industry with mountains of data is healthcare. Medical organizations need to manage the digital growth of data collected from MRIs, X-rays, different test results and other large files. These organizations also need to ensure data is instantly available while adhering to compliance and regulations, which makes object storage very attractive for convenience and decreased costs associated with management. Enterprises also expect to increase the levels of business intelligence they can leverage based on where reports are stored. For example, if a research team is mapping DNA sequences, they’ll want to store and analyze the data generated from this process, making object storage an ideal approach to house the information, given its scalability and accessibility.
Media and entertainment projects are also causing enterprises to reevaluate their storage approaches. As visuals have progressed from 1080p to 4K or 8K HD to 3D, movie and video game studios are looking to object storage as a cost-effective option. Additionally, since object storage tracks change history, it lends an extra layer of security, especially if the system is infected with ransomware and needs to roll back to a previous version of the object to retrieve the files.
We hear a lot of renewed interest from clients in exploring ways object storage can save them money and alleviate headaches, but they’re unsure of where to start. To make sure a foray into object storage goes off without a hitch, you have a few options. For one, we’d recommend tapping object storage for new projects that won’t require a lot of reliance on older systems. By starting a new island, you can work through any hiccups without affecting existing data. On the other hand, we see plenty of object storage opportunity through backups. This gives you a way to move important data to an easily scalable platform without affecting production environments.
Clayton Weise is the Director of Cloud Services for Key Information Systems, where he is responsible for designing, architecting, and implementing cloud solutions, managing production workloads and leveraging cloud resources in disaster recovery, clustering and hybrid (cloud and on-premises) infrastructure solutions.
Subscribe to Data Informed for the latest information and news on big data and analytics for the enterprise.