The stress of an unmaintained Apache Cassandra® solution was getting to the team of software engineers at the data-as-a-service (DaaS) company WellAware. They had applications to build, code to write, and more customers coming in. Something had to be done.

The industrial IoT technology and services company WellAware makes data, visibility and control possible in harsh environments like oilfields and wastewater facilities. They provide full stack services to companies that typically operate in challenging industrial environments with high temperatures, high pressure or offshore.

WellAware’s clients get cloud web products and mobile applications where users can view dashboards, get reports and manage alarms with real-time data from the field. This lets them improve their production on sites that aren’t easily accessible.

The team is passionate about making their clients’ jobs easier, and the full stack service makes WellAware an easy choice for a lot of clients. In the company’s early years, WellAware ran on a self-managed open source graph database. And as more clients were coming in, this proved to be a problem.

Hosted Cassandra solutions were hard to find

The problems for WellAware were increasing latency and slow scalability. It took too much time to read and write the high density data. And since WellAware usually stores client data for at least five years, they had to look into partitioning and sub tables as possible solutions. As their number of clients grew, they needed to change to avoid a loss of performance.

“It just wasn’t worth the time or more creative workarounds to keep our old database setup,” said Daniel Fake, Director of Cloud Software at WellAware.

“We evaluated different options. Some of our team had used [Apache Cassandra] before, so we looked for a hosted solution that would work well with our cloud hosting services on Amazon Web Services (AWS). Nothing really seemed very mature at the time. Then we decided to administer and maintain it ourselves, and that worked fine for a while.”

But then came a setback. WellAware’s sole expert with experience in Cassandra left the company. “We lost a lot of knowledge. The entire dev team did our best to step up and fill that role. But we lacked the deep knowledge needed to fine tune our Cassandra setup. And we didn’t have time to learn more,” Fake said, adding, “We didn’t even have time to run the repairs we knew how to run.” He said, “It was always at the back of our minds that this might come back to haunt us.”

Astra DB helped WellAware scale

With service-level agreements (SLA) at 99% uptime or better, it was only a matter of time before WellAware’s neglected Cassandra setup would become a critical issue for the company. For example, many of WellAware’s customers run pumps and tanks in hazardous environments like oil rigs, which makes that SLA of 99% uptime crucial to prevent dangerous situations.

Fake said, “If our clients go blind, there could be tank leaks, contamination or other critical assets issues that go unnoticed.” He explained that “there was a real fear that if there was a problem with Cassandra, we might not know what to do about it ourselves. We couldn’t risk it.”

So when he and the WellAware team became aware of Astra DB, the multi-cloud database-as-a-service (DBaaS) from DataStax, they quickly decided to make the switch.

The actual conversion from the application side only required minimal code changes in the connection setup. But it was also critical that WellAware maintained its 99% uptime promise during the migration.

Since WellAware had more than 600 gigabytes of data to move, the migration needed some time. This is where the DataStax zero downtime migration tool came in handy.

“It was a big migration for us, and we were worried how smooth it could go,” Fake Said. “But the DataStax team came prepared with a plan for us to discuss. Multiple engaged DataStax employees helped us and checked in with us through the whole process. It did take some time, but it went smoothly and without any impact on our business.”

WellAware has since gone on to make their tank and pump monitoring and control products available to even more customers through AWS Marketplace. The WellAware team knows now that they can scale up quickly when needed and even scale back down if another setback should hit the company.

Fake said, “We feel really good about the performance of Astra DB and how solid it is. We don’t want to have to plan a static cluster for peak loads and risk paying for more than we need. So that part of the DataStax service offering is also really nice.”

WellAware views DataStax as the stewards of Cassandra

Even though WellAware is indeed going places, they’re not planning to move away from DataStax any time soon. As new demands come up within WellAware and their products and new possibilities for using Cassandra in new ways emerge, Fake and his team will turn to DataStax for advice on where to go next. WellAware is currently eyeing the newer streaming solutions and will be considering those in the future.

“I feel like DataStax are stewards of Cassandra. They’re clearly the experts on running it. I know they’re very responsive and helpful. So I have a lot of trust in them to manage our database solution in the future.”

About Daniel Fake, Director of Cloud Software, and WellAware

Daniel Fake is an accomplished software engineer, with more than 15 years of experience creating innovative IoT systems for the energy and industrial sectors. He works to build and maintain cloud-based data platforms and elevate team performance. WellAware is a data-as-a-service company that helps make industrial businesses safer, more sustainable and more efficient through managed digital connections to hard-to-reach assets.

Resources

1. WellAware website
2. Amazon Web Services (AWS)
3. DataStax Astra DB
4. WellAware Tank level monitoring product on AWS
5. WellAware Pump monitoring product on AWS
6. 4 Steps to Migrate Live Data from Cassandra to Astra with Zero Downtime
7. Join our Discord: Fellowship of the (Cassandra) Rings