full
color
#d9d9d9
http://nerds.net/wp-content/themes/smartbox-installable/
http://nerds.net/
#08a1da
style2
(301) 622-7995

Case Study: Montgomery County Traffic

Case Study

Montgomery County Traffic Management Center

System & Network Administration

Montgomery County, Maryland with a population of 870,000, 3,000 miles of roadways, and a major mass transit system is part of the Washington, D.C. metropolitan area. To meet its transportation needs, the Advanced Transportation Management System (ATMS) was implemented. The ATMS is operated from the Transportation Management Center (TMC) and is based on over 20 years of experience in designing, constructing, and operating a real-time computerized signal system composed of over 780 traffic signals. ATMS features open architecture and is composed of multiple integrated subsystems that are managed and operated through common transportation workstations. ATMS is designed to provide control, monitoring, and traveler information in an effort to reduce traffic congestion, travel time, and accidents. Its emphasis is on safety and efficiency of mobility that include mode, route, and travel time choices.

At this state of the art center Traffic Engineers control the signals and systems that manage traffic in the County. Large amounts of data are collected and constantly analyzed to determine the best way to adjust traffic flow.

The Challenge

Extremely large volumes of data are constantly being generated, collected and stored in an extensive database and those data are continuously analyzed by the staff of the TMC to determine ways to optimize traffic flow and minimize delays, with emphasis on safety and efficiency of mobility that include mode, route, and travel time choices.

With so much irreplaceable information, data backups become extremely important. As the volume of data increased, the prior backup system, a DAT tape carousel that held 58 4mm DAT tapes with a capacity of 10-12gb each, became obsolete. The software was unstable and required purging and rebuilding the database monthly. The system required excessive yearly maintenance fees and the capacity was inadequate for the amount of data to be backed up. To alleviate this, multiple backups of different systems were scheduled each week, swapping out the tapes each time. The backups took well over 36 hours every week. Recovery required from 6 hours to a day or more, assuming the tapes were readable. Clearly there was a need for a better system of backups.

The Solution

The NERDS Group proposed a solution that eliminated tapes with their numerous problems and issues. The system that was implemented was a backup scheme to hard disk. The NERDS team built two custom backup servers, each with several large-capacity hard drives, and located one on-site and one off-site. The off-site system was accessed over a 1GB Fiber network link. The primary system was configured to pull backups from all servers, and this same server was used as a repository for all of the workstation backups and home directory folders. The contents of the primary backup server was then copied across the network to the remote backup server, where multiple generations of archival copies of the data are preserved.

Backups are conducted throughout each day, with the home folders backed up every 4 hours, and the servers backed up nightly. Each day at noon the remote server is updated. Using Rsync, this process is made extremely efficient because only data that has changed is transmitted. As a result, the process of backing up a relatively dynamic server with 10GB of data would take only minutes.

The Results

Since the switch to the new Backup System, the TMC has experienced additional growth and The NERDS Group has been there to add new hardware, software and support systems to facilitate that growth. In addition to substantial annual cost savings and reduced manpower requirements for the maintenance of the backup system, the reliability of the backups has increased exponentially, and the time needed to restore a file or entire computer system has been substantially reduced.

paged
Loading posts...
link_magnifier
#2c72b9
on
fadeInUp
loading
#2c72b9
off