The time required to analyze complex data sets was once measured in days, but due to the proliferation of faster processors and the advent of mass data storage, that expectation was reduced to minutes by the turn of the century. With the recent introduction of both hardware and software focused on Big Data applications, response times are now down to seconds, and in many cases, below the one second threshold.

When dealing with extensive datasets, both recorded and streaming, the desire remains the same, to generate actionable information, but there's a significant difference between taking action at a future point in time – tomorrow or next week – and taking action while an event is occurring. By implementing real-time Big Data analytics (RTBDA), key decisions, as well as actions, can be executed based on the analysis of both structured and unstructured data.

Big Data ComputingTo that end open-source software such as Apache Hadoop can be deployed on clusters of COTS servers in support of distributed processing applications that digest raw information stored in databases like Cassandra and data warehouses such as Hive, while in-memory analytics allow billions of records to be processed in a matter of seconds. These systems often reside within data centers which feature highly controlled environments, but in many applications, such as industrial automation or in military deployments, ruggedized servers designed to withstand shock, vibration and elevated temperatures are required.

  • 5U rugged rackmount storage server
  • Dual processor single board computer
  • 48 front access, hot swap 2.5? drives
  • Access control for multiple domains
  • BIOS modifications & revision control
  • Rugged, lightweight aluminum chassis
  • Long life cycle with 7+ year availability

Trenton's approach to supporting Big Data includes CPU processing power, combined with mass data storage capability and support for both GPU and FPGA computing. For example, the BXT7059 single board computer offers dual-processor, multi-core power to a variety of integrated systems that address the need to analyze massively parallel data streams.

Big Data Computing

  • Rugged, lightweight aluminum chassis
  • Dual or single-processor SHB options
  • 14-slot backplane supports PCIe Gen3
  • Up to 12 hot-swap, front-access drives
  • Up to 32 execution cores are available
  • Redundant hot-swap power supplies
  • BIOS modifications & revision control

Trenton Systems offers an array of PCI Express backplanes supporting both Gen2 & Gen3 transfer rates that are designed, manufactured, and supported in the USA.  In addition, our engineering team can design custom solutions for Big Data platform applications, including modifications of current designs, as well as ground up board designs based on application specific requirements.

December 22, 2016

Trenton Systems Inc.