There are a number of ways to work in a multi-cloud environment, but so far it looks like one provider – Flexify.io – offers a virtualisation layer on top of numerous cloud storage back ends.
The company claims it can make cloud supplier lock-in a thing of the past, and allow customers a simplified single point of access to a range of cloud storage providers that may differ in terms of capacity pricing, egress, performance, region etc.
Flexify also offers migration between clouds at costs that can come in below cloud providers’ egress charges.
That’s the offer proffered by Flexify.io. But where does the need for these services comes from and how does it work?
The backdrop comes from the increasing use of cloud compute and cloud storage, which are increasingly in the mix with on-premise IT assets. Just about all organisations use the cloud to some extent, with the most advanced use cases including the capability to burst out to cloud capacity to meet demand or for particular workloads.
But, not all clouds are created equally when it comes to storage. Some cost more than others to just store data, some are more pricy than others when it comes to egress costs, some may perform better than others, and others may simply be in the right place.
Those are some of the facts of life of cloud storage, which with the cloud more broadly, is increasingly in use alongside on-premise IT assets for most organisations. This means multi-cloud is not just a reality for many organisations, but a challenging issue when it comes to managing it.
So, what does Flexify.io bring?
Core to their offer is a single API-driven point of access to a virtualization layer with around 20 cloud storage providers behind it.
It is all object storage, and the providers behind the curtain include: Amazon S3, Azure, Google Cloud, Backblaze, Wasabi, Digital Ocean, Alibaba Cloud, Mail.ru Cloud, Yandex, Dell EMC ECS and Minio.
Application API calls land at Flexify using S3 commands. Behind that, Flexify handles routing to and from whichever storage providers the customer uses and in whichever combination is specified, for reads, writes, lists, with changes possible on the fly for reasons of performance, cost etc.
The Flexify virtualisation layer presents all data in a single bucket. It handles protocol conversion to allow access to supported clouds, but access for the customer is always via S3 while access methods are converted in the background.
While all that day-to-day storage I/O takes place, migrations can also be carried out between the cloud storage providers behind the Flexify veil. Application access is unaffected by background changes such as migration.
Benefits include being able to use a variety of clouds for reasons of performance, cost – such as egress, for example – and region. Data that might be kept close to compute in one cloud might be tiered off to less costly storage once processed. Meanwhile, data could be held in a particular geography for compliance reasons while access is via a single namespace.