Veeam reveals cloud storage use patterns: Azure and AWS for bulk, IBM has ardent fans, Google has … not much
Average cloudy data stash hits 15TB and and median value tripled to over 3TB
Data management software vendor Veeam has offered a snapshot (pardon the pun) of how its customers put different public clouds to work.
Senior vice president Anton Gostev's weekly missive to the Veeam community yesterday detailed how the November 2020 version 10 release of the vendor's software changed customer behaviour.
Gostev reported that of the five largest object storage repositories tended by Veeam, three were in Azure blob storage, on was in on-premises S3-compatible storage and the other was in Amazon S3. These individual object storage repositories were also five times larger than those Veeam had seen before, and ranged from from 1.8PB to 1PB.
The sizes dwindled from there, but remained significant – the next ten repositories had over 500TB of backups, the next 30 over 300TB, the next 60 over 200TB and the next 200 over 100Tb.
And while these sizes are individual object storage repositories, customers could very well keep multiple repositories.
But that's enough about the biggest repositories. Surely they were just outliers and on average, customers are sticking within range, right? Nope. Most customers stored more data than they did the prior year as averages grew by about 50 per cent to 15TB and median value tripled to over 3TB.
- This won’t hurt a bit, says Veeam, as it flags end of socket-based licensing
- New year, new OS: OneDrive support axed for old versions of Windows from 1 Jan 2022
- AWS has started upgrading the software behind S3 storage cloud
Most of Veeam's users employ Amazon and Azure for their cloud storage. The few that chose IBM tended to trust Big Blue with more data than those who work in other clouds. Google customers were hard to find, which Gostev attributed to Veeam's V11 release still being new.
Customers with over 100TB were split evenly between cloud and S3-compatible on-prem storage. Dell EMC's ECS had the most on-prem object storage deployments, but this year had competition from Hitachi HCP and Scality.
Aside from the hyperscalers, the top S3-compatible cloud storage was Wasabi, which was third overall and is five times larger than Blackbaze, which had a better year due to S3 Object Lock API support. Customers didn't fully utilize that feature, as less than half of object repositories had the option enabled. Gostev chalked that up to customers trying to save a quid or two.
Gostev also revealed that work on Veeam V12 is under way, and he predicted it will drive on-prem object storage adoption due to the new version having support for backing up directly to object storage. The version will also allow adding multiple buckets to Capacity Tier and ensure they don't go beyond scalability thresholds.®