#OlderDBFail: If Amazon and Pokemon Aren’t Immune, No One Is

You don’t always have to choose between the scalability of NoSQL and the accuracy of MySQL

High-value, high-transaction challenges aren’t just the domain of smaller players–some of the most resource-rich, technically advanced companies in the world have had problems, as evidenced recently by Amazon Prime Day, which was plagued with many problems that we predicted in a recent blog, Making Amazon Prime “Black Tuesday,” Not a Black Eye: “Anyone who stands on the stretch of Canal Street leading into the Holland Tunnel in New York City knows it’s almost impossible to avoid gridlock when thousands upon thousands of motorists cram into a 2.3-mile-wide island. Similarly, it’s quite the challenge for any e-tailer to prevent site slowdowns or crashes when they experience massive traffic spikes.”

We saw how tough that was last week, where, as Jason Ankeny reported, “Many Amazon Prime subscribers hoping to score Prime Day deals and discounts instead took to Twitter to complain about “add to cart fail” messages generated when they attempted to purchase sale-priced items. Some added the hashtag “#PrimeDayFail” to their tweets.” As we pointed out in our blog, having a database that can scale to the volumes of simultaneous transactions required would have alleviated traffic bottleneck and ensured pages loaded when shoppers expect them to, enhancing the customer experience and driving loyalty.

And consider Pokemon Go, which though it has since become a media sensation, was initially hit with some pretty serious technical problems. Despite a great team of veterans from Google, the Australian launch was immediately followed by a barrage of social media complaints over server outages. The fact that this wasn’t even a worldwide release further illustrates how tough it can be to scale applications to global demand.

The Clustrix way out

Fortunately, as SD Times recently pointed out, you don’t always have to choose between the scalability of NoSQL and the accuracy of MySQL. ClustrixDB is specifically designed for large, high-value transactional workloads. Where most relational databases, including Aurora and MySQL, are designed to Scale-Up–that is, increase performance by migrating the application to a more powerful server–ClustrixDB is the only relational database designed to Scale-Out–increase performance by simply adding commodity server nodes.

ClustrixDB maintains the accuracy of MySQL with ACID compliance, yet scales out the way other cloud applications scale, and can scale 10x and beyond Aurora (which runs on a single node). “For all acceptable measurements, ClustrixDB delivered the most transactions per second, while yielding a predictable performance curve and remaining under the 20-millisecond latency threshold,” said Mike Leone, Senior Analyst, ESG. “This threshold is crucial, as all it takes is a busy time of year to overload database capacity, sending response times through the roof—this can have an immediate impact on the end-user experience, customer loyalty and business revenues.”

As the worlds of Big Data and High-Value data continue to collide, companies no longer have to compromise by going with one or the other. Clustrix offers the benefits of technical simplicity, as customers only need to add instances (or commodity hardware) to the database cluster with a few clicks whenever they need capacity, while maintaining excellent (20-milliseconds-or-less) response times–where MySQL, Aurora and others begin to stall.

Not only is the price right….

Clustrix also provides both ease of maintenance and stability, offering a drop-in MySQL replacement that requires no application modification. Unlike MySQL databases that have been sharded or had read slaves applied to them, it has no single point of failure and each node is a peer to the others and can maintain database performance when one (or more) fail.

And, the cost benefits of ClustrixDB cannot be over-emphasized. With ClustrixDB, customers can “right-size” their database in true utility fashion. There’s no limit to how much ClustrixDB users can expand, and they can just as easily “flex” back down and subtract instances after peak season. ClustrixDB also works with the MySQL procedures with which most DBAs are already familiar, so there’s no need to replatform, hire new tech folks to maintain the database, or waste valuable time with the aforementioned technical workarounds.

For those customers dealing with high volumes of high-value transactions, Amazon Aurora will trigger hidden fees once you hit the the “fast lane” and a moderately successful company will pay twice what they anticipate in transaction fees. Clustrix customers, on the other hand, always know what they’re spending—their tab only fluctuates when they “Flex” up and down by adding and subtracting instances to match ebbing and flowing demand. There’s no penalty for success with ClustrixDB.

No room for compromise with today’s applications

Most of today’s web/cloud applications require high value/high volume transaction processing with low latency. This need for accuracy combined with scalability means that traditional solutions won’t be an option. We’ve already “cracked the code” with ClustrixDB, the first relational database that scales out so that you can flex performance indefinitely by merely adding or subtracting nodes, while ensuring data reliability at all levels of performance.

Any way you look at it—whether from a technical, staffing, cost, or business strategy perspective—Clustrix just plain makes sense. As MakeMyTrip, India’s leading online travel site, Rakuten, one of Japan’s largest e-commerce sites, and numerous other companies will attest, Clustrix enables you to sail through performance challenges that quite often plague even well-funded initiatives that rely on more traditional database approaches.