cutimage - Fotolia
Maxwell Health was wrestling with a homegrown backup system for its MongoDB databases that could only do full, continuous hourly backups, until the company switched to Amazon Web Services cloud backup with Datos IO RecoverX.
The software-as-a-service healthcare company had built a microservices platform deployed on MongoDB and natively on the AWS cloud platform. It was a costly configuration because Maxwell Health was backing up 250 GB every hour with a free backup tool built into MongoDB. It got the job done, but not without problems.
"We were unhappy with our solution for a long time," said Ashley Penney, director of infrastructure at Boston-based Maxwell Health. "It was clear it was not going to scale and it was not robust. We could only do a full backup every hour and it was very, very expensive. If we needed to do a restore, we would have to restore the entire thing."
Maxwell Health provides software as a service for employee health benefits. As part of its product development, the company does continuous integration and development on multiple production and test database clusters. It was backing up to 11 TB of data daily and the hourly backups usually were held for a week and gradually reduced to a copy stored in Amazon Glacier.
"We would keep three copies a day after two weeks, and after a month we moved the copies to long-term storage," Penney said. "It was relatively easy, but it was a pain. And there were times we needed to restore and we could be waiting all day for the restore. Glacier makes it very easy to move data to the cloud, but it's not easy to get it back. And if I needed to restore, I would have to restore the whole copy."
Before implementing Datos IO RecoverX backup and recovery software, Maxwell Health tried to implement an enterprise-grade backup system provided by MongoDB, but that turned out to be complex and even more costly.
Ashley Penneydirector of infrastructure, Maxwell Health
"It's expensive because it requires you to build even more infrastructure," Penney said. "You have to build a second MongoDB cluster to stage the backup. We tried, but it turned out to be a bad experience all around. It was not stable at all. The clusters would not work. They would go down or we would get error messages so we would scrap them and start over again. We would have to recreate the cluster from scratch."
Penney found Datos IO backup software via a Google search. The company provides a cloud-scale, application-centric data management platform that uses an elastic architecture with parallel streaming and application-consistent backups. It works on a private cloud, hybrid cloud or public cloud.
The Datos IO RecoverX is aimed at application architects, database administrators and development teams. It can manage, protect and recover third-party platforms and internet of things data that reside locally and on SaaS platforms. It works with scale-out databases, including Amazon DynamoDB, Apache Cassandra, DataStax Enterprise, Google Bigtable, MongoDB, Redis and SQLite.
"Datos IO built us an image that we launched in the cloud and from our side we were able to test it," Penney said. "They are doing a lot of things that are cutting edge and they were really quick. Now with RecoverX, we are doing backups every hour but they are incremental. RecoverX deduplicates data so now I'm storing 13.2 TB of 83 TB in the cloud. It deduplicates 99% of the data.
"We've gone from storing 110 TB to storing 183 GB. That is a huge savings for us," Penney said.
How to choose a software and cloud provider
Understanding data dedupe in the cloud
Check out our cloud-based backup guide