Cloud storage: the evolution of 2022

In 2022, storage administrators must continue to find ways to make their users work from home and further integrate cloud and on -premises infrastructures. For their part, suppliers will try to meet these needs by developing more hybrid solutions.

“There’s a reluctance between putting everything in the cloud or keeping everything on site. Moving data from one to another is expensive. Deploying hybrid IT that addresses this problem makes sense,” he said. review comment by Ray Lucchesi, from Silverton Consulting.

Another challenge is careful financial monitoring to find a balance between the expenditures that DevOps teams want to make on Kubernetes solutions and the hidden costs of cloud storage.

LeMagIT spoke with Ray Lucchesi and other storage experts to find out what trends they see developing in 2022.

The cloud will expand on on-premise storage offerings

The current context is that more and more companies are thinking of an infrastructure that encompasses on-premises and cloud-based resources for storage. One of the new reasons they cited for no longer being 100% dependent on an on-premises data center is the fear of component shortages that will prevent them from buying the storage array they want.

In 2022, vendors will try to encourage these companies to use self-administered cloud storage services to facilitate the transition. These providers include hyperscalers such as AWS, Microsoft Azure, and Google Cloud Platform, as well as infrastructure vendors that offer pay-as-you-go online services, including Dell and HPE. .

“The cloud is not really cheaper if you have a lot of data. Just the process of buying and managing is easier. »

Dave RaffoAnalyst, Evaluator Group

“With the cloud, vendors no longer need to sell you equipment and companies no longer need to buy storage arrays. On the other hand, the cloud isn’t really cheaper if you have a lot of data. Only the purchasing and management process is easier, ”said analyst Dave Raffo of the consultancy Evaluator Group.

Cloud storage providers are trying to make their online services more attractive. Last year, AWS launched self-administered file services that match common on-premises technologies, such as OpenZFS and Netapp OnTap. These new services make it possible to approach on -premises file systems to transfer current workloads to the cloud and from there connect to other AWS services.

According to Dave Raffo, the step that will make such online storage services a success is, among major public cloud companies, the provision of managed disaster recovery services, i.e., Activity DRPs) turnkey. AWS started the trend by launching the AWS Elastic Disaster Recovery service in 2021. Rival GCP (Google Cloud Platform) boasts the services of Actifio, a PRA company acquired by GCP in 2020.

“Traditional backup vendors already offer a SaaS console, or use the cloud as a target resource to host backups. From now on, the people of the cloud will offer themselves to do this backup, ”Dave Raffo reviews.

Dave Raffo sees another lever for online storage services managed by the cloud provider itself: block storage, or SAN. Such services should target database projects.

Services available to meet intermediate needs

The problem with online storage offers is that their very clear features limit use cases. For example, either we store data on S3 Glacier at the best price, but there is a barrier to waiting a long time when we want to recover it, or we store this same data on S3 plain and simple, without such barriers, but more expensive.

IDC analyst Andrew Smith believes these offerings will branch out into adjacent offers, bringing in more intermediaries. He cited Amazon’s latest S3 Glacier Instant Retrieval service, a minor version of S3 Glacier’s long-term storage service that allows data to be retrieved in fractions of a second compared to a few minutes normally. The price, however, is higher. Similarly, S3 Glacier is also available as the S3 Glacier Deep Archive, which takes 12 hours to retrieve data, but is still cheaper than the main service.

“It’s about giving businesses more options. Options that, from a price point of view, will vary only a few tenths per GB stored. But when you know that companies are likely to store more data for longer periods of time, you understand that a few dimes of a penny can have huge consequences on overall cost depending on your usage, ”he said. .

According to him, AWS competitors will mimic its versions. He also thinks freemium offerings – free to some extent – will expand to cover a narrower range of needs.

The Kubernetes are spreading now

According to Dave Raffo, more vendors will offer Kubernetes-compatible storage, which retains container data after the container is closed. Containers and their Kubernetes orchestrator are required for running new so-called cloud-native applications. These applications are business of DevOps, which urges CIOs to provide them with appropriate storage solutions, not legacy SAN / NAS storage arrays.

“In addition, developers working with Kubernetes often work in the cloud. So all storage vendors are doing their best to support Kubernetes on -premises, but also in the cloud,” he said.

Ray Lucchesi added that the open source aspect of Kubernetes and its portability means that this system is now a standard for hybrid infrastructures. He is therefore convinced that its use will increase further in the coming years, as will the need for storage administrators to host data on these containers.

Leave a Comment