The massive and sudden shift to remote work in 2020 that resulted from the pandemic pushed many organizations to ramp up their cloud strategy. The pandemic “validated the cloud’s value proposition,” according to the analysts at Gartner, and nearly 70 percent of organizations using cloud services today plan to increase their cloud spending as a result.
For many organizations, the switch was made quickly just to get something in place—with some taking almost a lift-and-shift approach to the cloud. That will work for the short term, but now it’s time to look at the long haul of the cloud transformation journey. And, since data is a key asset for corporations during this journey, this is where fast object storage can play a key role.
2020 Fast-Tracked the Adoption of Cloud Services
Spending on cloud increased in 2020 and is anticipated to continue in 2021, with the proportion of IT spending that’s shifted to the cloud accelerating. Cloud is projected to make up 14.2 percent of the total global enterprise IT spending market in 2024. That’s up from 9.1 percent in 2020. According to Gartner, overall cloud services spending globally is expected to grow by nearly 25 percent in the next four years, to a cumulative $438 billion.
The almost overnight need for large-scale remote work capabilities favored a short-term cloud strategy. Enterprises have quickly embraced SaaS—tools like Salesforce.com and Intuit are a few popular examples. In some cases, they may not even realize that the use of these SaaS apps means they’re in the cloud. They’ve also been quick to embrace collaboration tools like Zoom, Slack and GoToMeeting.
It’s the longer-term cloud strategy that many companies are still just beginning to approach. However, there are challenges to overcome in the long term. One such challenge is cloud vendor lock-in. While many companies are talking about a hybrid or multi cloud approach, this isn’t always actually happening, largely due to issues of vendor lock-in.
Another challenge is that apps need to be refactored. Refactoring is the process of running your applications on the infrastructure of your cloud provider—that is, you have to completely re-architect your applications to better fit and optimize for the new cloud environment. This can be tricky because you have to ensure that, while making application code changes, you don’t affect the external behavior of the app.
Storage and the Cloud: Where Fast Object Storage Comes In
IDC forecasts that more than 500 million cloud-native apps and services will be developed and deployed by 2023. That figure equals the total number of apps created in the last 40 years. Since data and users are now everywhere, these applications need to be flexible and agile in terms of where they can be deployed—data center, cloud or edge. What’s more, organizations need to make use of the mountains of data coming from these apps. Fast object storage is the enabler to getting the value from all that data.
Over the last 10 years, people have tended to think of object storage as the way to deal with
“secondary” data—backups and long-term archives, to be specific. This is based on a somewhat outdated notion that, while object storage could provide advantages in “scale” (growth of capacity), it had the connotation that performance would be insufficient for higher-performance applications.
In reality, object storage can provide very high levels of throughput (fast delivery of data per second), whether in the cloud or on-premises. This is the performance metric that is the most important for Big Data payloads, like those seen in cloud applications (images, videos, audio files and documents). The truth is that, for a wide range of applications managing unstructured data, object storage is indeed their primary storage solution.
Breaking Down Siloed Data Tiers
It’s an accepted fact that continually growing data will soon become unmanageable. With that as a given, you also should accept that spreading all that data across multiple clouds (storage tiers in silos) is also unmanageable. There are two dimensions of unmanageability: the volume of data and the number of places you keep it.
As stated earlier, traditionally object storage was considered adequate for only a subset of data. It was seen as great for Big Data that you barely touch, but not for the data you use frequently. However, that is no longer true as fast object storage can eliminate the problem of siloed data storage tiers, thereby breaking barriers and misperceptions.
An optimal solution is a single tier or model of storage that is fast enough to be considered the main tier for (at least) 80 percent of your data. That’s what object storage can now offer: a single tier that you don’t have to worry isn’t big enough, fast enough or cost-effective enough to hold everything—because it is.
Preparing for Today and Beyond
Now that the dust kicked up by the rapid shift to remote work has settled, organizations are beginning to take stock of what’s been put in place and whether it will serve their needs long-term. Immediate cloud “fixes” now must be re-evaluated within the scope of a future-focused cloud strategy. The profusion of apps brings with it a proliferation of data that organizations need to capture and make use of. And while object storage was once viewed as a slow solution for rarely touched data, fast object storage is now busting that myth and offering a large yet manageable and cost-effective method for modern data storage needs.
Paul Speciale is chief product officer at Scality.