Mason to Receive $2.3 Million Computer Security Grant
Posted: March 23, 2007 at 1:00 am, Last Updated: November 30, -0001 at 12:00 am
Photo by Evan Cantwell
Many people have experienced the frustration of having a computer crash because of a virus or software failure. Usually, the only option is to rebuild the system from scratch and inevitably lose all files stored on the computer.
But Anup Ghosh, Mason professor and chief scientist at the Center for Secure Information Systems in the Volgenau School of Information Technology and Engineering, is doing his part to improve the way we respond to failures and attacks on our computers and servers.
Ghosh will serve as the principal investigator (PI) on a Multidisciplinary University Research Initiative (MURI) award of more than $4.8 million from the U.S. Department of Defense. As the lead institution, Mason will receive $2.3 million over five years. Other collaborators on the project are researchers at Columbia University and Penn State University.
The project, entitled “Autonomic Recovery of Enterprise-Wide Systems after Attack or Failure with Forward Correction,” will develop a system of detecting attacks, corruptions and failures in enterprise-wide servers and client workstations. The systems will then self-regenerate to the last known “good state” to recover data that otherwise would be lost.
“Enterprise computing is a critical resource in the era of network-centric warfare,” says Ghosh. “Logistics, transportation, intelligence, command and control and even modern combat systems all depend on the correct operation of enterprise computer systems. Enterprise computer systems need to be not only highly available, but also highly resistant to attack.”
The technique being developed by Ghosh and the other researchers will allow for imperfect software systems to be deployed, but by providing autonomic recovery (the ability to fix itself) and regeneration, these systems will recover easily and be more resistant after attack. The team adopted a “health care” model for computing systems, in which failing systems are brought back to health while other systems provide their services.
By taking snapshots of the system at scheduled intervals, the system maintains a record of the last known good state. The transaction-based model will enable consistent and forward recovery with correction by quarantining and resetting lost files to their pre-corruption state.
Although this software is being developed for use by the U.S. government, Ghosh plans to have it licensed to allow for further development and commercial use.
The highly competitive MURI award to the team headed by Ghosh was one of 36 awards to academic institutions.
Other PIs involved in the project include professors Sushil Jajodia (Mason), Angelos Keromytis (Columbia), Salvatore Stolfo (Columbia), Jason Nieh (Columbia) and Peng Liu (Penn State), who received his PhD at Mason under Jajodia.