It seems hard to believe, but we’re approaching 6 months since the launch of the project we’re internally calling Object Storage Gen2. In 2025, between January and June, we launched Gen2 in 8 global regions, with additional regions coming in the second half of this year. A retrospective is a great way to share some of the feedback from customers who have been helping to steer our journey so far, and to provide some insight into what’s coming next.
As a general-purpose Object Storage service, we support a wide range of customers and applications. Akamai is known for its expertise in content delivery, so it’s no surprise that a common use case for Object Storage is serving as the storage origin for media and content distribution with our media and security products. But as the Akamai Cloud Computing platform grows, we’re also seeing rapid growth in cloud native use cases in areas like logging, observability, machine learning, and data analytics.
A common recurring theme across industries and applications is the need to deliver larger, higher-performance buckets globally.
Although we have already been successful in partnering with customers to support events like the Super Bowl, regional elections, and game launches, our pursuit of higher performance and improved scalability never stops. Data underpins the foundation of all applications. Whether it’s increasing user engagement or expanding datasets, increases in capacity and performance allow customers to scale without worrying about data placement to maintain consistent application performance. Higher performance and scale also open the door to consolidation of a wider range of applications across teams, so they can benefit from the operational, cost, and ecosystem advantages that an S3-compatible object storage service can provide.
From the Linode acquisition in 2022, we’ve made the financial and engineering investments necessary to grow the number of regions that support Object Storage from 4 to 22.
Our teams have also done a great job at consistently delivering higher performance and scale to our customers.
| Scope | 2022 | 2023 | 2025 | |
| Max Requests Per Second | Bucket | 750 | 5,000 | 20,000 |
| Max Capacity | Bucket | 10TB | 1PB | 5PB |
| Max Objects | Bucket | 50M | 1B | 10B |
| Default Quota | Per Endpoint | 5TB | 100TB | 500TB |
As applications store more content, accessed at higher rates with shorter lifecycles, the next broad area that our customers are asking us to partner with them on is in the area of visibility and access management. Improved metrics, audit trail, and control over who can access data and role restrictions are areas that we can address as part of the core service to reduce the effort for our customers to meet their business objectives. We’re also working closely with customers to help optimize the indexing of data using applications like Elasticsearch and Quickwit, as well as managing large datasets using Apache Iceberg.
I do try to engage directly with as many of our customers as possible. For those I have spoken to directly, hopefully you see how your input is shaping our product. If we have not spoken yet, all feedback is always welcome and appreciated. I’m always looking for new insights and perspectives to help influence where we go next.
Tags
