ScaleArc, a provider of database load balancing software, has announced support for automatic cache invalidation, providing a transparent ACID-compliant caching mechanism for dynamic data. As a key feature in ScaleArc’s database load balancing software, automatic cache invalidation increases website and application performance by making it safe to cache data that frequently changes, such as shopping cart and user profile information for e-commerce applications.
The ScaleArc software tracks data changes to invalidate a cache entry, ensuring that outdated data is not served. ScaleArc also provides the ability to handle more workload, reducing page download time and increasing website performance, while protecting data.
Cross-Platform Database Support
ScaleArc’s database load balancing software is available for Oracle and MySQL, as well as SQL Server database environments. “Our entire platform is a universal SQL state engine and SQL rules engine that we have tuned for Oracle, MySQL, and Microsoft SQL Server RDBMSs. And so every time we bring new technology to market, it universally applies to those three database types,” said Michelle McLean, vice president of marketing at ScaleArc.
Cross-platform support for all three of these relational database management systems has been a key differentiator for the company, said Varun Singh, founder and CTO. “There is one seamless way to provide operational excellence with all these databases, whether it comes to how to do caching, analytics, or how to do high availability, as well as the load balancing functions that we provide. The aim is to make sure we provide a common set of features and functions across all three of these databases.”
Patented Technology
According to ScaleArc, the auto cache invalidation capability was evaluated by a leading e-commerce company. The company tested the feature across two query sizes, measuring the query-per-second (QPS) rate and response time both with and without ScaleArc’s database load balancing software. Throughout the testing, the company observed that response time with ScaleArc’s software increased 6x to 12x, depending on the query response size.
“We were granted a patent on our caching technology a little more than a year ago. That patent was primarily around how we do transparent caching for all of these databases without changing anything in the application layer,” explained Singh.
“The secret to that is that it is based on a NoSQL technology,” said Singh. “We are the first NoSQL technology that is transparent and looks like SQL to an application layer. We are able to provide the kind of performance that you get from a NoSQL cache and we are able to provide the kind of scalability that you get from a NoSQL cache - without any of work and effort required to achieve that.”
Key Use Cases for Automatic Cache Invalidation
Key use cases for auto cache invalidation include shopping cart data, user profile data, and financial data. For example, shopping cart data, despite being unique for each user, can now be cached since the auto cache invalidation feature will ensure the shopping cart will reflect changed items immediately.
In addition, for user profile databases which need to have high uptime because they are validating users before any transactions can be performed, data tracking by the auto cache invalidation feature ensures user details are up to date while offloading the database from processing highly repetitive queries.
Auto cache invalidation also is beneficial for financial data including e-commerce sites where users can buy and sell items in an auction and require the auction price and associated data, to be fetched and updated frequently. Other frequently changing financial data, such as stock and option prices or banking information that must always be consistent, can also take advantage of the offload offered by ScaleArc’s cache, the company says.
“The automatic cache invalidation is useful in environments where the data changes frequently and it needs to be transactionally accurate for the applications. Both of those are pretty broad categories,” said Singh.
“Until now there has not been an ACID-compliant cache that has been available that can automatically invalidate very granularly the entries for each one of those individual things,” he said. “The performance is equivalent to most NoSQL technology that has been created. It brings about the best of both worlds and puts it in a unique place where it can do the kind of things that neither the NoSQL side nor the SQL side could have done in the past.”
For more information on evaluating ScaleArc’s software, click here.