How Qumulo Helped AWS Customer Reduce Latency by Shrinking Distance Data Had to Travel

By Gokul Kuppuraj, cloud architect, Qumulo, Inc.

Learn how Qumulo, Inc. enabled a large university to lessen latency and increase efficiency.

When leveraging the cloud, it can become challenging to run applications that require low latency and high performance while still maintaining reasonable costs. This is especially true when the cloud provider’s closest supported region is over a thousand miles away.

One of the company’s higher education customers in Arizona was using Amazon FSx for their enterprise file share workloads. Managing 100+ FSx shares with their existing solution was proving to be too slow and expensive. The largest challenge was the closest AWS region for FSx was over a 1,000 miles away in Oregon.

The distance between FSx and the customer’s workload created a performance issue when latency became unacceptably high. High latency creates bottlenecks in the workflow, even if it’s only a handful of milliseconds here and there. This customer moves large amounts of data, so having it stored closer makes all the difference.

Local zones for AWS are a type of infrastructure deployment that places compute, storage, database, and other select AWS services close to large populations and industry centers in order to serve specific regions. They are designed to help bring data closer to their customers and reduce latency. Not all local zones are set up to handle all AWS services, however.

Bringing data closer to home
The customer needed a solution to reduce latency. As an existing company’s customer for their multi-petabyte research projects, the university knew the firm could help.

3x improved throughput and reduced latency
By running Qumulo in an AWS local zone closer to Arizona, the customer was able to deploy a cost-effective solution with increased performance, and gained real-time visibility into all their data on a granular level. In the end, the customer was able to consolidate its file share workflow and saw 3X improved throughput and lower latency compared to AWS FSx running in the US-West-2 (Oregon) region.

The customer moved their most demanding internal customers to Qumulo first. Since this initial deployment was so successful, plans are underway to move the next phase of unstructured data workloads to Qumulo on AWS.

Scale anywhere with Qumulo
The company makes it simple to manage petabytes of unstructured data anywhere and forever. In the cloud, the firm’s modern scale-out file system enables high performance workloads and provides intelligent data services needed to meet SLAs.

Qumulo on AWS supports all regions except China, including GovCloud, Local Zones, and Outposts. By leveraging the infrastructure building blocks available on AWS, the firm can provide the same unstructured data solution everywhere. Applications can benefit from low latency by running in the local zone closest to the end user.

With Qumulo on AWS Local Zone, customers get:

• Enterprise unstructured data solution
• Reduced latency between clients and applications
• Increased throughput
• Local data residency needs met
• Cost-effective solution

In summary, latency is reduced, and customers also see cost savings and a boost in throughput.

Resource:
If your organization is struggling with latency issues and can’t get the level of service you need, email aws@qumulo.com to learn more about Qumulo on AWS.