Follow Datanami:
December 15, 2014

Aerospike Enhances In-Memory NoSQL Database

— Aerospike Inc., the world’s fastest database, today unveiled new features and enhancements to its open-source, flash-optimized, in-memory NoSQL database solidifying the company’s lead in fueling the next generation of real-time, context-driven applications.

With the availability of powerful new clients, easier installation and deployment, storage and performance improvements, enterprise security enhancements, Hadoop integration and numerous community contributions, Aerospike continues to meet and exceed developers’ current and future needs. Its unmatched speed, scale and simplicity make it possible for organizations of all sizes to innovate with new applications and rapidly grow new businesses that drive top-line improvements, while accruing massive bottom-line savings.

“There is enormous potential for today’s businesses to harness the power of big data with real-time applications, but the challenge is finding breakthrough database technologies that aren’t confined by traditional designs,” said David Floyer, resident CTO, Wikibon. “Aerospike’s next-generation NoSQL database brings price and performance to new levels, fueling disruptive and transformative new applications.”

“Aerospike continues to blaze the way with innovations to its NoSQL database technology far outpacing competitors,” said Brian Bulkowski, co-founder and CTO, Aerospike. “Both disruptive startups and established enterprises choose Aerospike to power a new class of real-time, data-driven applications that aren’t technically or economically feasible with other database technologies.”

Unlike first-generation NoSQL databases and traditional in-memory technologies, Aerospike is self-managing and scales on a handful of servers instead of hundreds of servers. With recent enhancements to the open-source Aerospike, code-based developers benefit from:

  • More flexible development
    • Upgrades for Java, Node.js, .NET / C#, and C Clients.
    • New Python, PHP, Go clients and a beta Ruby client.
    • The ability to set consistency on a per-transaction basis to support multiple applications and ensure applications that tolerate eventual consistency can benefit from lower latency.
  • Easier installation and deployment
    • Docker containers and Vagrant images simplify deployment on laptops, bare metal or the cloud.
    • Google Click-to-Deploy expedites creation of a two-node Aerospike cluster on Google Cloud Platform.
    • Aerospike AMI and enhanced operational guide simplify deployment on Amazon Web Services.
  • Storage and performance improvements
    • A new disk write cache allows reads to be done from recently written data. For this common access pattern, one customer deployment resulted in a 40 percent reduction in SSD utilization while only increasing RAM use by one percent.
    • New jemalloc memory allocation system greatly reduces memory fragmentation for pure DRAM configurations.
    • Ability to configure either time-based or size-based disk flush using the O_SYNC device flag to avoid file system and device buffers.
    • New hot-swap SSD storage capability allows disks to be added or removed on the fly.
  • Enterprise security enhancements
    • Role-based authentication and authorization capabilities give users read, read/write and administrative control.
  • Hadoop integration
    • InputFormat integration eliminates ETL to HDFS, enabling Hadoop tools to analyze data stored in Aerospike.
    • Hadoop Split and Query support Aerospike’s patented Indexed MapReduce capability so Hadoop jobs can operate on a targeted subset of data in Aerospike, instead of the typical HDFS pattern of analyzing large data lakes. For some uses, this results in 100 times performance gains.
    • OutputFormat integration allows petabyte Hadoop jobs results to be immediately injected into Aerospike.
  • Robust community contributions to Aerospike Labs
    • More than 33 community contributions and counting, within months of going open source. Aerospike Labs links to connectors, tools, applications and code examples that accelerate development. Recent contributions include Apache Spark integration enabling Spark to efficiently read data from Aerospike.
    • Additional contributions include a JSON document manipulation library, a real-time messaging application, Scala wrapper, Ganglia monitoring module, Express session store integration, URL shorteners, simple real-time recommendation engine in Spring, Java Eclipse integration, Magento caching module and Apache Storm integration.
Datanami