We are excited to bring Transform 2022 back in-person July 19 and virtually July 20 - 28. Join AI and data leaders for insightful talks and exciting networking opportunities. Register today!
Amazon Web Services (AWS) yesterday announced that Amazon DocumentDB (with MongoDB compatibility) now supports what the company is calling Global Clusters, a new feature that enables enterprises to deploy a cluster that spans multiple AWS regions. In effect, this can help companies more quickly recover from regionwide outages — even in less than a minute.
This version of Amazon DocumentDB is a fully managed NoSQL document database service that supports MongoDB workloads and makes it easy to store, query, and index JSON data. With this new feature, Global Clusters can now replicate such data to clusters across regions with latencies of less than one second, using dedicated infrastructure with little to no impact on workload performance. It was previously possible to deploy an Amazon DocumentDB cluster with instances that could span three Availability Zones, but this didn’t protect against regional outages.
The announcement follows other recent news from AWS, including the general availability of both Amazon Location Services and Redshift ML.
The main benefit of support for Global Clusters is in the event of an outage. If one occurs, the new feature allows one of the secondary regions to be promoted to full read/write capabilities in under a minute, according to AWS.
Customers can have up to five secondary regions, and each secondary region can have up to 16 replica instances. What’s more, the number and type of instances in the secondary clusters doesn’t need to match the primary, enabling enterprises to scale as needed. Regardless of the volume of data, AWS says scaling instances in Amazon DocumentDB takes less than 10 minutes.
So why should a company’s data strategy consider disaster recovery? For one thing, this goes beyond a hardware failure. If there is a regionwide failure, which has happened before, it can knock enterprises offline for that time period. The recent fire at a OVH datacenter, for example, showed the need to consider multiple geographic regions when looking at cloud services. Overall, the Global Clusters functionality can step in and enable enterprises to recreate a global database in a different region without data loss.
The addition of Global Clusters fills a gap in disaster recovery for the MongoDB-compatible database service. While the ability to deploy an Amazon DocumentDB cluster with instances that could span three Availability Zones has been available, this didn’t help in the event of a regional outage.
Keep it local
In addition to outage recovery, AWS identified low-latency reads across various local regions as a secondary use case for the new feature. This essentially brings data closer to applications in different global regions, which has significant benefits for any applications that are globally distributed. Specifically, serving users data locally improves the customer experience. When people abandon shopping carts because a page took too long to load, for example, it underlines the importance of being able to replicate across regions and keep data “local.”
By using global clusters to replicate data to other regions, users can also read data from the secondary cluster in the region closest to them. When use cases have a high read-to-write ratio, this also has the advantage of serving reads locally with low latency, leaving the primary cluster for writes.
In a blog post, Amazon describes in detail how to set this option up, as well as failover steps in the event of a regional outage.
VentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative enterprise technology and transact. Learn more about membership.