|
ImageHighlight |
|
|
Gem Data
article 2
|
|
ThisIssue |
|
|
RSSFeed  |
|
|
|
|
|
NewsLinks |
|
|
|
|
|
CommunityCorner |
|
|
|
Site Highlight: Embry-Riddle Aeronautical University |
|
New equipment serves data and images to a broad spectrum of users, perhaps you?
Christopher G. Herbster, Embry-Riddle Aeronautical
University, Daytona Beach, Florida
The Applied Meteorology
Program, within the Applied Aviation Sciences Department, at the Daytona Beach
Campus of Embry-Riddle Aeronautical University (ERAU-DB) has used funds from
the 2008 Unidata Equipment Award program to provide a cornerstone to a new
design in distributed computing for our external and internal web and data
server architectures. The funds provided from this award were used to purchase
a Sun Microsystems SPARC Enterprise T5240 Server. This server provides a
staggering combination of transaction and memory resources, with 64 1.2GHz cores
that can utilize 64 GB of memory. While the original proposal was going to
provide for a system with “only” 32 cores and 32 GB of memory, the Applied
Aviation Sciences Department was able to provide the difference between the
award funds and the purchase price of this robust server (~$2,000) during the
“Matching Grant Program” that Sun typically offers to educational institutions
in the month of June. Aside from the incredible transaction rate this server
is capable of, it was chosen because of some other desired attributes that the
Solaris operating system and Sun hardware provide. The system has the ability
to self diagnose hardware problems and then exclude those components from the
continued operations of the server. Many of the hardware elements can be
replaced without powering down the rest of the server, thus helping to provide
a minimum of down time for our services.
This image from Aug 21, 1115 UTC, shows the full domain of the ERAU CONUS image, with hurricane Bill and the NHC forecast track in a visible channel sector of the image, while the bulk of the U.S. is covered by IR data.
|
The Solaris operating system
allows for an interesting capability to create virtual servers within the core
of the system. A series of individual servers operate within a virtual
structure, such that “restarting the server” is now a restart of the virtual
system (a program) from the parent operating system structure. These virtual
servers are found on the Internet as a number of addresses on the Applied Meteorology’s
ever growing Internet accessible web page resources (http://wx.erau.edu/data/, http://fltwx.db.erau.edu/, http://wxkiosk.erau.edu/, and http://wx.erau.edu/erau_sat/). The flight weather data server provides the
flight planning resources that are utilized for all of the pre-flight weather
briefings our (roughly 1500) students in the ERAU-DB Aeronautical Science
curriculum perform. Annually our flight line performs roughly 150,000 hours of
aircraft operations, or an average of roughly 20 aircraft in the air for any
hour in the 20-hour work day the flight line operates. Another virtualization
serves as the “Wx Kiosk” portal. This server is the core of a passenger, employee
and visitor display at a weather and flight information kiosk (with a
half-dozen walk-up stations) at the Daytona
Beach International Airport (KDAB; about 600,000 passengers per year). All of these web servers rely
heavily on a variety of GEMPAK shell scripts and McIDAS generated products to
complete their missions. more
The newly-purchased hardware will allow ERAU to: provide the IDV internally in our classrooms and to student workstations; share forecasts made by the NWS as well as those made on our own cluster. This will facilitate information dissemination.
There are:
- 1200 students in the Aeronautical Science (Professional Pilot) curriculum, who rely on our web services for pre-flight planning and classroom exercises.
- 400 students in our Air Traffic Management program training to be FAA controllers.
- 130 students in the Applied Meteorology BS program who rely on the server for classwork.
- 120 in the Safety Science program
- 140 students in our new Homeland Security BS program (who take at least one meteorology course, though many complete the Aviation Weather Minor).
- 25% of our entire student body of 4500 undergraduates (1,125) takes a meteorology class every semester.
This year we are beginning a new Area of Concentration in our Master of Science in Aeronautics in Aviation Meteorology. We expect that the server will be used by this group for analysis of weather information in their research. |
| |
The Data Series: Gem Data |
Unidata now relays GEM data within the IDD topology.
The Global Environmental Multiscale (GEM) model is an integrated forecasting and data assimilation system developed jointly by the Meteorological Research Division and the Canadian Meteorological Centre of Environment Canada.
The GEM model is currently operational for deterministic and probabilistic (ensemble) global data assimilation and medium-range forecasting, deterministic regional data assimilation and short-range forecasting. Furthermore higher-resolution limited-area forecasts are produced overnight and are available to the operational forecasters. A growing number of meteorological applications are now either based on or use the GEM model such as air quality forecasting, wave forecasting, dynamic extended-range forecasting on monthly to seasonal time scales.
The essence of the approach is to develop a single highly efficient model that can be reconfigured at run time to either run globally at uniform-resolution or to run with variable resolution over the globe but such that high resolution is focused over an area of interest or with high uniform resolution over a limited area. The data assimilation for deterministic forecasting uses the variational approaches in either its 3D or 4D version; for probabilistic forecasting it uses the Ensemble Kalman Filter approach.
The GEM model has been developed to meet the operational weather forecasting needs of Canada for the coming years. In the future they will include nowcasting for cities and airports, regional ensemble forecasting, sea-ice assimilation, coupled ocean and atmospheric assimilation and various environmental parameters coupled to the atmosphere. The limited-area version is also being developed as the next version of the Canadian Regional Climate model.
 |
This is the CMC-GEM North American domain at 60KM resolution.
Fields displayed are the 35 m/s (meters per second) iso-surface and MSLP (mean sea level pressure) contoured and filled. |
In April 2002 the Canadian Meteorological Center (CMC) and Unidata announced the availability of model forecasts from the operational Global Environmental Multiscale (GEM) regional model, and until May 2009 the CMC delivered all GEM output "point-to-point." Unidata worked collaboratively with CMC to provide the LDM technology for use in the point-to-point data delivery, but recently we revisited the project to propose making the data available via the IDD. We're happy to report that CMC was amenable to the change and the rest is history. This distribution method is mutually beneficial, easing the use of CMC's bandwidth, as well as being beneficial for the Unidata community, because it provides multiple points of access.
Heaviest users of the IDD-delivered GEM data are universities in the northern tier of the community.
|
Software Updates |
GEMPAK: A bugfix update to the 5.11.4 source release was made the last week of August. The latest information on the N-AWIPS to AWIPS II migration can be found at http://www.unidata.ucar.edu/software/gempak/nawipsmigration/.
IDV developers released v. 2.7 u2 in late August. Some new features included in this release are: Parallel Data Reading and Rendering; First guess grid for Objective Analysis; Export data to netCDF.
McIDAS X, In 2009 and beyond, McIDAS-X software updates will be made via software addenda (including online documentation updates) rather than via formal annual upgrades each summer. This step is being taken so that the McIDAS programming, testing, and documentation efforts can focus on the next generation McIDAS, McIDAS-V.
RAMADDA, The Unidata Program Center is pleased to announce the first release of the RAMADDA Data Repository. RAMADDA brings together data serving, content management, publishing and collaboration facilities. See http://www.unidata.ucar.edu/software/ramadda/index.html for more information.
THREDDS, In late August, THREDDS Developers released version 4.0 of the THREDDS Data Server. Probably the most significant new feature is the addition of an OGC WMS service, thanks to collaborators Jon Blower (University of Reading UK) and Pauline Mak (Australian Research Collaboration Service).
|
News Briefs |
 Training Workshop
Mid-August saw the end of 2009's Training Workshop. Although the sessions entail a great deal of extra work for Unidata's developers, and planning for them is a non-trivial as well, the entire staff is energized by having the participants visiting here in the Program Center. As reported in the August e-letter there were 91 participants. In addition to the Peruvian attendees shown in the picture on the right, participants came from Canada, Colombia, Germany, and Saudi Arabia as well as the U.S.
The next Training Workshop sessions will be scheduled for fall, 2010. To accommodate differing schedules, we alternate workshop times from mid- to late summer, to mid- to late fall.
Follow Us:
WAYS YOU CAN GET US: RSS | Newsletters and Alerts |
 Facebook  Twitter |
Please send comments to support-eletter@unidata.ucar.edu
The CommuniteE-letter is produced by editor, Jo Hansen, and production manager, Tina Campbell |
|
|
|
