Oracle has released a major new version of its NoSQL implementation, promising that it will allow easier management of 'high-velocity transactional data' such as that generated by social media and smart sensors.
The company unveiled Oracle NoSQL Database 2.0 on Monday. The database uses Oracle's Berkeley DB storage engine and, according to the firm, is more tightly integrated with Oracle's Database and Hadoop environments than earlier iterations.
"Oracle NoSQL Database 2.0 provides developers with that 'last-mile' infrastructure to enable crucial real-time interaction between an organisation's applications and its customers," Oracle development chief Vaishnavi Sashikanth said in a statement. "It provides a key mechanism for enterprises to maximise the value of big data."
According to the company, the new version offers better performance "with near linear scalability" and latency below 5ms. There is a new API for C developers, and new APIs for automatic serialisation and handling large objects.
Also on Monday, Oracle announced its Big Data Appliance X3-2 and a new version of the Big Data Connectors software suite.
The upgraded appliance is based on Intel's eight-core Xeon E5-2600 processors. It provides 288 CPU cores — 33-percent up on the previous version — and 1.1TB of main memory. Oracle also claims it requires "up to 30 percent" less power and cooling.
The new Big Data Connectors release sees the Oracle SQL Connector for Hadoop supported within the Data Integrator Application Adapter, and also provides "transparent access" to the Hive Query language from the R statistical computing language.