Rafter Absorbs an Influx of E-Commerce Traffic With ClustrixDB

School is nearly back in session, and that means students will scour the Internet in droves to get the textbooks they need for their studies.

It’s no secret that the cost of higher education can be significant. Rafter, Inc., a leading e-commerce retailer, understands that perfectly and as a result, offers textbook rentals that can save students as much as 80 percent off retail prices.

Already in the company’s brief history (incorporated in 2006), business grew 700 percent between 2009 and 2010. Between 2010 and 2012, the company’s sales tripled.

Those numbers would be welcomed by any executive. But Rafter quickly realized that in order to accommodate the increase in traffic, its technological infrastructure needed to be upgraded. As it stood at the time, the company’s infrastructure was built on MySQL, with a master/slave configuration online casino in dual data centers.

Rafter’s complex setup created multiple points of failure, particularly at the database layer. It took a lot of work to maintain, and to make significant changes, it required extended systemwide downtime. The company also had problems with redundancy, and because Rafter used an open source platform, it was virtually impossible to ensure a prompt response to software issues that arose.

To mitigate those problems, Rafter deciding to make the switch to ClustrixDB, a scale-out NewSQL cloud database. In doing so, the company got a database that promised high availability, scalability, reduced overhead and a low total cost of ownership. The switch was uncomplicated and quick, and Rafter was able to accommodate a growing business model with ease.

“The support around [our implementation] was amazing, and there was never a question on the right thing to do,” said Michael Farace, director of systems and infrastructure operations. “We wouldn’t be able to grow our business as easily without ClustrixDB.”

After migrating to ClustrixDB, Rafter experienced an unexpected 100 percent increase in database load the following year. The NewSQL database was able to absorb all of that traffic without a hitch.