The Intersection of Cloud Repatriation and Active Archiving
It has been widely reported in the trade press that some organizations are questioning the cloud-first philosophy that emerged over the past decade. This concern has given rise to workloads being moved from the cloud back in-house, known as cloud repatriation. With an increasing focus on lowering costs, improving cost predictability, maximizing data accessibility, and raising the level of control of data security, the cloud repatriation trend is gaining momentum.
There are many drivers behind this trend. Some organizations have faced staggering bills for cloud usage. Some have even quit the cloud entirely due to rising costs. Others want more predictability in their storage costs and find that the public cloud does not always provide that. More than a few organizations face data sovereignty and control challenges, either due to regulatory limitations or heightened risk – they need at least certain portions of their data to reside close to hand.
But it isn’t a case of abandoning the public cloud and bringing everything back into the enterprise data center. There is no doubt that the public cloud will continue to play a major role in the world of storage. What is needed is a measured evaluation of what workloads ideally belong where. Some workloads should stay in the cloud, and certain kinds of new workloads should also be sent to the public cloud. However, some workloads should be retrieved from the cloud.
Repatriation, then, doesn’t mean turning back the clock to the infrastructure you had prior to large-scale cloud adoption. It means building your own private cloud to supplement the public cloud. It also entails recognizing the values and aspects of private cloud-based storage that make sense, add value, reduce costs, or increase accessibility.
Build Your Own Infrastructure – But Get Help
While many workloads will continue to operate in the public cloud, many others need to be hosted within your own private cloud. For some, hot data and mission-critical workloads that aren’t dependent on cloud-based services will be brought in-house and served to users on superfast flash systems or based wholly in memory. And as the amount of cold data expands, much of it is probably best housed in an in-house active archive rather than in the cloud. Thus, organizations need, once again, to how best to invest in their internal infrastructure.
A private storage clouds may offer the right balance of performance, cost, and simplicity at scale. There are no transaction fees, storage fees, or egress fees. Cost containment is assured as hardware costs continue to drop faster than public cloud prices. REST interfaces provide easy access and efficiency by virtue of a multi-tiered, low-overhead approach that can be an efficient way to store data. A private storage cloud also offers features for durability, availability, and data protection.
But organizations, these days, don’t have to do it all themselves. They can and should take advantage of the many resellers, managed service providers, and vendors that have experience in building as-a-Service solutions or deliver managed service solutions that allow you to tailor your own active cloud storage infrastructure. This approach makes it possible to establish a cloud that you fully own – and one that is comparable or even less than the cost of a public cloud. This becomes particularly important to consider in light of the anticipated growth in data over the coming decade.
Building the Right Storage for Tomorrow
The storage infrastructure of the future will comprise tape, disk drives, solid-state disks, and other forms of storage. As there are dramatic cost differences between these elements, it becomes vital to catalog and map your data to where it needs to be to achieve the right mix of performance and cost efficiency. In addition, it must be possible to seamlessly migrate data between these different media and systems.
That takes thinking like a hyperscaler and deploying a data management infrastructure that allows you to move data meaningfully across data types.
You can find out more about emerging use cases for active archives by listening to the recordings of the 2023 Active Archive Virtual Conference
