|
Real-time
applications are indispensable for conducting research and business in
government, industry, and academic organizations. Recently, real-time
applications with security requirements increasingly emerged in
large-scale distributed systems such as Grids. However, the complexities
and specialties of diverse security mechanisms dissuade users from
employing existing security services for their applications. To
effectively tackle this problem, in this paper we propose a security
middleware (SMW) model from which security-sensitive real-time
applications are enabled to exploit a variety of security services to
enhance the trustworthy executions of the applications. A quality of
security control manager (QSCM), a centerpiece of the SMW model, has
been designed and implemented to achieve a flexible trade-off between
overheads caused by security services and system performance, especially
under situations where available resources are dynamically changing and
insufficient. A security-aware scheduling mechanism, which plays an
important role in QSCM, is capable of maximizing quality of security for
real-time applications running in distributed systems as large-scale as
Grids. Our empirical studies based on real world traces from a
supercomputing center demonstratively show that the proposed model can
significantly improve the performance of Grids in terms of both security
and schedulability.
|