topMenuBar
topMenuBar

Information Management and Technology

Whether through implementation of numerical models and/or design of modern real-time environmental monitoring systems, Woods Hole Group is heavily involved in technology development and applications. Our technologies for data management and visualization cut across the Oceanographic, Coastal, and Environmental core technical teams, allowing Woods Hole Group to deliver modern solutions to our customer base. Specific capabilities include:

  • Database Design and Implementation
  • Geographic Information Systems (GIS), Graphics and Mapping
  • Civil/Mechanical CADD
  • Software Development
  • Modeling



Snapshot:

Data Hosting and
Web Display


Woods Hole Group provides privately hosted, secured websites on mirrored servers and with redundant fiber-optic connections. We provide worldwide access to ...[+]

Parallel Cluster
Computing


Woods Hole Group engineers built a high-performance 96-node parallel computing cluster to help with the demanding calculations required by high-resolution...[+]

Data Hosting and Web Display


Woods Hole Group provides privately hosted, secured websites on mirrored servers and with redundant fiber-optic connections. We provide worldwide access to client data via an easy-to-use customizable interface that provides "information at a glance" or in-depth scientific analysis. While such services might reside on the Cloud, often our customers have highly confidential and specialized data that must be hosted in a private, secured, redundant system with maximum guaranteed uptime. For these customers, Woods Hole Group can:

  • Receive the data
  • Perform QA/QC on transmitted data sets
  • Convert raw data into usable files
  • Provide analytics and visualizations
  • Host the data on a secure, redundant private site
  • Present data results in a web format
  • Archive the data

Parallel Cluster Computing


Woods Hole Group engineers built a high-performance 96-node parallel computing cluster to help with the demanding calculations required by high-resolution models. The cluster is built from a dozen Linux servers, each with a pair of 64-bit quad-core Intel processors and 8 gigabytes of RAM. A switched gigabit network with separate networks for inter-process communication and storage connects the servers. The system has a theoretical peak performance of 768 GFLOPS (theoretically multiply two numbers together 768 billion times in a second), and allows our modelers to run simulations of far greater resolution in a fraction of the time than previously. We are currently adding another 72 processors that will increase computational capacity to a teraflop, over a trillion floating-point calculations per second.


download as PDF

Download Full Project Description