Parstream

An Analytical Database for Big Data

ParStream offers a revolutionary approach to high-performance data analysis. It addresses the problems arising from rapidly increasing data volumes in modern business, as well as scientific application scenarios, by employing innovative algorithms and patented index structures.

Use Cases

Analyzing Big Data is getting increasingly important for companies in all industries. Examples therefore are:

  • Web Analytics – analyze billions of records within milliseconds
  • Social Media – learn more about your users
  • Market Research – ParStream enables fast and flexible combinations of many columns
  • Climate Research – Standard SQL features and custom query nodes provide maximum flexibility and speed for applications such as Geo-Clustering
  • Health Care and Insurance – Simultaneous data import and querying allows comparison of current & historical data
  • Telco – Quality of service monitoring and billing
  • Mining – filter and query on vast amounts of geodata
  • Smart metering – analyze meter data across the customer base

Benefits:

  • Ultrafast
    get analytical results in fractions of a second
  • Scalable
    easily store and analyse billions of records
  • Unique
    perform analytics that were previously not feasible
  • Realtime
    simultaneously import and analyse data
  • Efficient
    use the parallel processing power of GPUs
  • Cost Effective
    save on infrastructure costs

Setbacks in Current Database Architecture

The figure above shows a performance comparison of ParStream and DBMS X as platform for an online search engine for travel offerings.
Current databases are not engineered for mass data, but rather for small data volumes up to 100 million records. Today’s databases use outdated, 20-30 year old architectures and the data and index structures are not constructed for efficient analysis of such data volumes. And, because these databases employ sequential algorithms, they are not able to exploit the potential of parallel hardware.

Supported Platforms

Currently ParStream is available on a number of Linux Distributions including RedHat Enterprise Linux and SUSE Enterprise Linux running on X86_64 CPUs. On request, ParStream will be ported to other platforms.

ParStream has full support for GPU acceleration on NVIDIA Fermi Cards.