DataStax has introduced the latest version of its database platform based on Apache Cassandra. With version 4.0, the NoSQL database adds a new in-memory option as well as enterprise search enhancements for faster data processing for high-performance financial, telecom and e-commerce applications where speed is paramount.
By providing integration of real-time information with a production-certified version of Apache Cassandra, DataStax aims to help organizations build scalable online applications that leverage hot data in real-time, in the context of their critical business application. According to the vendor, DataStax customers currently include eBay, Netflix, Adobe, Constant Contact, Ooyala, and more than 20 companies in the Fortune 100.
DataStax: focusing on performance for online applications
“We are solely targeted on online applications, not data warehousing but the operational online database, and our goal over the last couple of releases has been to supply an enterprise NoSQL platform that is really geared for heavy duty production environments - something that supplies all the maturity and niceties that people are used to from the relational database world, but with all the power and flexibility of NoSQL bundled in,” noted Robin Schumacher, vice president, products, DataStax.
“What you are going to see from us in the next couple of quarters, starting with DataStax Enterprise 4.0 is a major focus on performance – and that is going to be across the entire product line, whether it is open source or commercial. With DataStax 4.0, what we are bringing to the table to enhance the already very fast performance we offer with Apache Cassandra and DataStax Enterprise is the new in-memory option.”
The new in-memory capability is intended for use cases where speed is of the essence, and offers “a very nice piece of flexibility that allows our customer base to really dial that performance knob and be able to target their data to whatever storage mechanism they need in terms of performance,” said Schumacher, depending on whether the data is cold or hot. Administrators can flexibly assign data to in-memory objects, traditional spinning disks or SSDs all in the same database cluster, making performance optimization easier.
“The great thing is that the in-memory tables are transparent to applications and to developers, meaning that when you create a new table all you have to say is that it is an in-memory table and nothing else. In essence, we bring all of the goodness of Cassandra that people already enjoy - the flexible data model, no single point of failure, the linear scale performance, the multi-data center support - all of those things now are brought to an in-memory database which people really haven’t had before. It is very easy to use and very powerful, and is ideal for use cases that require very low latency-style requests where speed is of the essence,” said Schumacher. In-house benchmarking indicates that under the proper conditions customers will realize anywhere from a 10x-to100x performance improvement, he added.
DataStax Enterprise 4.0 features certified version of Cassandra 2.0
The new release has a DataStax-certified version of Cassandra 2.0, which allows customers to enjoy all the new features and benefits of Cassandra 2.0 such as new developer enhancements. “What we offer is a complete certification process where we take Apache Cassandra, put it through our own certification routines of very heavy QA work, real-world benchmarking, defect resolution – and ensure that that version is ready and certified,” said Schumacher.
Lastly, DataStax 4.0 provides new search features to help developers build applications faster while enhanced internal cluster communications deliver faster search operations, even for thousands of concurrent requests. In DataStax Enteprise 4.0, said Schumacher, the company has updated the version of Apache Solr to 4.6 and which supplies a number of new developer features that make it much easier to develop search applications, and also created a new communication layer that makes it much faster for nodes to communicate with one another.
DataStax OpsCenter 4.1 supplies capacity management capabilities and visual monitoring of database clusters
Concurrent with the release of DataStax Enterprise 4.0 is a new release of OpsCenter, which allows users to visually manage their clusters whether they are on premise or in the cloud, said Schumacher. “We have added some really nice new features to OpsCenter, including the monitoring of those new in-memory tables and we have also enhanced the capacity planning capabilities inside the tool.”
Image courtesy of Shutterstock