Speaker
D. Skow
(FERMILAB)
Description
There have been a number of efforts to develop use cases for the Grid
to guide development and useability testing. This talk examines the
value of "mis-use cases" for guiding the development of operational
controls and error handling. A couple of the more common current
network attack patterns will be extrapolated to a global Grid
environment. The talk will walk through the various activities
necessary for incident response and recovery and strive to be
technology neutral.
Gedanken incident response exercises are being discussed among the HEP
PKI infrastructure specialists, but a systems-wide approach to the
issues and necessary tools is needed. Determining scope of incidents,
performing forensics and containing the spread requires a much more
distributed approach than our previous experiences. A new set of tools
and communication patterns are likely to be needed. This talk will be
aimed at applications and middleware developers as well as operations
teams for grids.
As time allows, the talk will survey current grid testbed middleware,
identify the current control points and responsibilities and suggest
places where extensions or modifications would be beneficial.
Primary author
D. Skow
(FERMILAB)
