USA flag logo/image

An Official Website of the United States Government

Automated Wide-Area Network Configuration from High-Level Specifications

Award Information

Agency:
Department of Defense
Branch:
Defense Advanced Research Projects Agency
Award ID:
77887
Program Year/Program:
2009 / STTR
Agency Tracking Number:
06ST1-0016
Solicitation Year:
N/A
Solicitation Topic Code:
N/A
Solicitation Number:
N/A
Small Business Information
Galois, Inc.
421 SW Sixth Avenue Suite 300 Portland, OR 97204-1662
View profile »
Woman-Owned: No
Minority-Owned: No
HUBZone-Owned: No
 
Phase 2
Fiscal Year: 2009
Title: Automated Wide-Area Network Configuration from High-Level Specifications
Agency / Branch: DOD / DARPA
Contract: W31P4Q-08-C-0356
Award Amount: $749,653.00
 

Abstract:

Wide-area networks are essential to productivity in commercial enterprises and military organizations. Reliability is difficult to achieve because wide-area networks, such as the Internet, are built from many autonomously configured component networks. Today there exists a large gap between the intended design of a network and the implementation of that design. This is because complex network policies must be expressed in low-level, device-dependent configurations. Routes are computed by composing configuration instructions across the entire network; thus, simple logical errors, or typing mistakes could inadvertently compromise global connectivity. This work bridges the gap between routing-policy specification and implementation for Internet interdomain routing. We propose a new paradigm in network configuration: one in which network operators can declare a policy specification in a high-level language, leaving automated tools to check the policy for errors and distribute configuration instructions consistently and correctly among network devices. Our approach is to design a family of policy languages to support different configuration needs and will incorporate recent research advances that help prevent anomalies in routing protocols. Our toolset will automate the configuration process, so we can also develop tools that dynamically modify network configuration in response to network measurements for purposes of traffic engineering.

Principal Investigator:

Peter White
Project Manager
5036266616
peter@galois.com

Business Contact:

John Launchbury
CEO
5036266616
john@galois.com
Small Business Information at Submission:

Galois, Inc.
12725 SW Millikan Way Suite 290 Beaverton, OR 97005

EIN/Tax ID: 931278540
DUNS: N/A
Number of Employees:
Woman-Owned: No
Minority-Owned: No
HUBZone-Owned: No
Research Institution Information:
Yale University
Department of Computer Science
P.O. Box 208285
New Haven, CT 6520
Contact: Hudak
Contact Phone: (203) 432-1235
RI Type: Nonprofit college or university