The challenges posed by the relentless creation of data in the enterprise are well known by now. Aside from finding a place to keep it all, and then quickly retrieving key files when needed, runaway data growth creates problems for security, compliance and a host of other requirements.
These problems are usually exacerbated by traditional hierarchical storage architectures. Whether it is the folders, files and types of basic file storage of the tracks and sectors found in block systems, the issue is the same: too much data that requires too much overhead to keep it all organized.
With object storage, however, there is no need for the vast metadata needed to locate a given object, even in storage pools that scale to exabyte proportions. This is the main reason many large-scale organizations, such as cloud providers and hyperscale enterprises, have switched to object storage as their primary data preservation solution.
According to IDC, the volume of digital data will see 42 percent compound annual growth for the rest of the decade. Already, we are starting to see global data loads in the zetabytes — 1 million petabytes – putting pressure on today’s storage systems to not only scale to extreme levels, but to do so seamlessly. Object storage fulfills this need easily, but it also brings a number of other key attributes to the table, both for managed service providers (MSPs) and systems integrators working to bring their clients into the 21st century.
Securing Storage
Security concerns are growing at a pace even with the rise of data volumes. As recent headlines can attest, ransomware and other types of cybercrime are evolving at a rapid clip. Fortunately, most object storage offerings have a number of built-in features to combat these threats and concerns.
For one thing, most object storage solutions are equipped with storage-level or object-level data encryption. Data security shouldn’t stop at the network perimeter with your firewall, and object storage has additional layers of security and controls which can help. As well, they utilize authentication solutions like the lightweight directory access protocol (LDAP) or Active Directory (AD) integration, which can be tied to object-level access lists (ACLs) for a more granular approach to security.
In addition, most object protocols utilize RESTful API calls and enable per-transaction tracking/auditing on access requests to data. This provides the ability to track historical changes to the data or to mark certain objects as immutable. All of this creates an additional security layer to ward against threats like ransomware, while at the same time simplifying the restoration process should data be destroyed. Object solutions also provide an added layer of protection that some experts call the “air gap,” caused by the fact that storage is not accessed natively by a server or like typical filesystem, but is accessed only via the object API. All of these features combined make object storage an ideal solution for large volumes of high-value, unstructured data.
Cutting Costs
Even the most optimal solution is of little use if it blows the budget. In this light, object storage is becoming increasingly valuable because it provides the most cost-conscious solution for other money-saving initiatives like backup-as-a-service and disaster-recovery-as-a-service.
Since its inception, object storage has made dramatic gains in cost-effectiveness through its ability to utilize commodity hardware and increasingly sophisticated software. This gives it a distinct advantage over traditional integrated storage systems in which hardware and software are tightly bound. As more object solutions are released into the channel, we can expect prices to drop even further, which in turn creates new opportunities for service providers, system integrators and enterprise users to shed bulky, aging tape and disk systems. And the economic advantages do not apply to large enterprises alone; many small and mid-sized organizations will see dramatic savings through object storage as well.
Setting the Stage for Infinite Expansion
Rampant data creation is the main reason data volumes are expanding so rapidly, but the flip side of this coin is the near absence of data removal. Enterprises are a lot like people, after all. Few individuals actively delete photos and other data from their phones until, of course, they run out of storage space. And even then, the typical solution is to delete just enough to get by or to simply buy more storage.
In the enterprise, of course, the need to preserve data is often fueled by retention and compliance regulations, which themselves can be inconsistent or confusing. Certain data types, for example, may need to be maintained for longer periods than others, but then these requirements may shift from one jurisdiction to another. As data infrastructure evolves to a global footprint, most organizations err on the side of caution by simply saving all data all the time. This, in turn, drives demand of flexible, scalable archive solutions that cater to the flattened architecture of object storage rather than the stacked approach of NAS or SAN storage.
Broadening Access
As a centralized storage architecture, object storage provides the most cost-effective means of enabling data access from pretty much anywhere. But users should be aware that object storage is not intended to support instant access to any and all data. For this reason, it probably isn’t the best solution for critical applications running in real-time, such as autonomous vehicles or in-memory databases where data is being analyzed on a continual basis in order to support rapid, complex decision-making.
Instead, object solutions provide optimal support for data repositories that require infrequent access to critical data under tight security protocols. This can include law enforcement services, particularly high-data applications like bodycam video, as well as healthcare imaging and record-keeping.
This is why two of the most important criteria to help evaluate an object storage solution are how often data is accessed and exactly what type of data it is. Most cloud providers will often use these metrics to calculate their monthly fees, which gives object storage a further advantage when it comes to important but infrequently used data. In general, if your archives are approaching the “80/20 rule” then it’s probably time to consider object storage as a scalable, secure solution for unstructured data.
One thing is clear: data generation will not slow down in the near future and will likely continue to grow at an exponential rate. For this reason, the enterprise has no choice but to seek increasingly innovative storage solutions. Object storage is on a steady improvement track in terms of both cost and security, which should ease the storage burdens that all organizations face in an increasingly digital economy.
