Technology is constantly in flux. To stay on top of cloud storage trends, businesses considering adding cloud storage need accurate information about this movement. Below, companies in the cloud and data storage space share the cloud storage trends they are seeing so you can be more informed. Take a look!
Emerging Technology Influencing Cloud Storage Trends This Summer
Artificial Intelligence
The rapid growth of Artificial Intelligence (AI) across industries is influencing the cloud storage trends this summer. Technologies like Machine Learning (ML) and Generative AI (GenAI) depend on the ability to access and manipulate large datasets in the cloud, requiring large capacity, storage flexibility and great throughput. Cloud Storage solutions are therefore evolving rapidly driven by the need for better performance, security and cost-efficiency. Cloud-native and the increasing use of serverless computing in event-driven applications and microservices are also shaping the storage technology landscape. Many Businesses are adopting a hybrid and multi-cloud setups are developing strategies that offer flexibility and improve resilience, while avoiding a vendor lock-in by the large public cloud providers. – Efrain Ruh, AIOps Expert and CTO at Digitate
AI/ML is a primary driver for moving data to the cloud from its initial location, which might initially reside on-premises or another cloud resource. After moving the data, the next step is to transform it into a format that can be consumed for a particular workload. Often, the initial transformation is followed by a second or third transformation to meet security or compliance requirements.
Transformation is a euphemism for using compute resources. Generally, compute is the most expensive resource in the cloud, but that must be calculations regarding the data life cycle. For instance, it may be cheaper to retransform data again, rather than pay storage fees. It is entirely situationally dependent.
Each transformation will require an appropriate data life cycle policy to be applied to it to minimize costs. A common requirement is that the data be moved from colder to warmer storage (and back again) based on the needs of the AI/ML workload so it can be used for future training activities or additional transformations. – David Christian, Global Migration Lead at DataArt
Cloud-first Policy Adoption
Leading into the summer season, ‘cloud-first’ has become a widely-adopted rule for companies that want to compete in the data-driven economy. With data only increasing in volume, the massive cost savings afforded by the cloud make it impossible for many organizations to opt for on-premises data centers. Of course, there are always exceptions to the rule. Still, today, cloud providers offer the security, flexibility, and often even the data residency requirements needed for any company’s unique circumstances.
In today’s collaborative data environment, where data is shared between departments, team members, and even satellite organizations, an on-premises solution can’t match the scalability and efficiency of the cloud. Disparate data was once a major stumbling block. Still, the ability for data platforms to do away with siloed and connected data from wherever it is has been addressed effectively by the cloud in a way that on-premises solutions can’t. – Sharad Varshney, CEO at OvalEdge
Software-defined Storage
One technology that’s not necessarily new but is becoming increasingly important is software-defined storage. Think of it this way: You’ve got lots of data in storage—a video archive, for example—that you don’t access often. There’s always an access pattern that emerges around this kind of data. But sometimes, those patterns change depending on what type of content people need at the moment. Traditionally, archivists notice that change and manually move some of the archival data into hotter storage so people can access it faster and cheaper. Software-defined storage builds frameworks to automate that process through scripting or AI to optimize for cost and performance. – Majed Alhajry, technology, business process, and software development leader at MASV
Security Concerns Shaping Organization Approach to Cloud Storage
Client Misconfigurations
A significant proportion of cloud security breaches are due to client misconfigurations, which are often driven by a lack of cloud expertise. That’s why some cloud providers have moved to the shared responsibility model. This model stakes out a middle ground between cloud providers dictating everything you can and cannot do, on one hand, and leaving customers to fend for themselves on the other. Shared responsibility means cloud providers implement sensible defaults—such as strong password enforcement or ensuring new storage buckets aren’t made public by default—while allowing customers the flexibility to configure their storage to suit specific use cases. – Majed Alhajry, technology, business process, and software development leader at MASV
Cloud Governance
Cloud governance is always an important element in any enterprise cloud implementation. Using cloud-native tools, Config or SecurityHub in AWS, Defender for Cloud in Azure, and Security Command Center in GCP allows you to see the security state of all storage repositories. It reports on questions like: Is the repository encrypted? Is the repository encrypted in a cost-efficient way? Does the repository have a life-cycle policy assigned to it? Does the repository restrict access from the Internet or internally? Are the policies that allow access to the repository the least privileged? Are permanent access keypairs disallowed or severely restricted?
Finding a repository that is out of compliance will mean scheduling it for a change to meet compliance needs. Creating new, out-of-compliance repositories is generally prohibited by policy. – David Christian, Global Migration Lead at DataArt
The Role Environmental and Sustainability Plays in Cloud Storage Trends
Optimizing Resource Use
Environmental and sustainability considerations in cloud storage often focus on optimizing resource use. Cloud providers therefore allow users to select the best type of storage and deliver technological solutions that allow them to move data to more cost-effective platforms with ease, reducing not only their carbon footprint but also costs. – Efrain Ruh, CTO / Cloud Management Professional at Digitate
Centralization of Cloud Storage
There’s a lot of greenwashing in the cloud storage space, but there is merit to some of it. For example, the centralization of cloud storage is one of its most important sustainability features: If everyone in the cloud decided to build their own data centers, the amount of space and other resources required would far exceed what they’re using in the cloud. The capacity of the cloud is also higher due to economies of scale, which means you can store more gigabytes per cubic foot, which means less need for cooling, silicon, and other resources. Some public clouds have even started using underwater data centers, which use ocean water as a cooling method and require far less power. – Majed Alhajry, technology, business process, and software development leader at MASV
Efficient Allocation of Resources
According to reports, 60% of all corporate data is currently stored in a public cloud. Cloud providers have economies of scale within their data centers that simply cannot be matched by corporate data centers. In a data center, the tendency is to leave the compute resources, bare metal, and virtual machines on 7×24 in case they might be needed to process data. In the cloud, from a customer’s point of view, when the data needs to be processed in some way, the compute is enabled, the data is processed, and the compute is turned off. From the cloud provider point-of-view, what is actually happening is the compute is reallocated to other customers, but the overall carbon footprint is reduced globally due to a more efficient allocation of resources. – David Christian, Global Migration Lead at DataArt
The Influence of Remote Work and Hybrid Work on Cloud Storage Trends
Flexible and Scaleable Storage
The shift towards remote and hybrid work models has pushed for more flexible and scalable cloud storage solutions. As teams continue to work remotely, there is a higher reliance on collaborations tools like MS Teams, Slack, Google Workplace, etc, requiring a robust cloud storage solution. Remote work often introduces security vulnerabilities, making data protection a top-level concern. Having confidential organization data being accessed from multiple locations and devices increases the risk of an attack or a breach. Effective backup and recovery capabilities are also crucial to minimize risks on a hybrid work model. – Efrain Ruh, CTO / Cloud Management Professional at Digitate
Expanded Geographic of Workforce
Hybrid and remote work makes cloud storage a necessity, especially if you have a geographically spread-out workforce. That geographic spread can create significant expenses for on-prem organizations that need employees to access storage from anywhere with low latency. You also can’t provision on demand with on-prem storage—you have to provision for the worst-case scenario, just in case—so companies constantly overpay for capacity they don’t usually need. The economies of scale built into cloud storage suit hybrid work models because they allow organizations to scale up and down quickly without requiring significant CapEx. – Majed Alhajry, technology, business process, and software development leader at MASV
Recent Unexpected Uses for Cloud Storage
Storage as a Service
Several use cases for cloud storage have emerged beyond traditional data storage. One example is Storage as a Service (STaaS), a solution that organizations are starting to adopt to reduce complexity and increase efficiency through a consumption-based as-a-service model with increased levels of automation. – Efrain Ruh, CTO / Cloud Management Professional at Digitate
AI Payloads and Training Data
The cloud is very well suited for AI payloads and hosting AI training data use cases, which require rapid access to data and large amounts of sequential reads. Cloud storage is well-suited, efficient, and cheap for these use cases. It gets very expensive to have a training data corpus stored on prem—those drives must be spinning all the time to provide on-demand access, even though you’re not training your model at all times. With hot storage in the cloud, you get that access on demand, and access to that data is usually free, so you’re only paying for the storage element. This applies across most industries. – Majed Alhajry, technology, business process, and software development leader at MASV
Consolidating Compute Resources
We’ve been seeing organizations that initially took a multi-cloud approach, begin to reconsider and consolidate into a single cloud. Overcoming data gravity is real, egressing data between clouds or even regions within a cloud is more expensive than originally calculated. Putting all compute resources in local proximity to data repositories has been a trend recently because it is more efficient. – David Christian, Global Migration Lead at DataArt