Current Status as of June 2002

	Protocols for Supporting Real Time Resource Management
	Protocols and Infrastructure for Supporting Fault Tolerance
	Methodology and Protocols for Supporting Security
	Protocols for Supporting Scalable Real-Time Multicasts
	Protocols for Supporting Mobile Wireless Environments
	Bio-Networking Architectures
	Middleware for Supporting Application Programs with End-to-end QoS
	Tools for Supporting Network Simulation/Emulation

Protocols for Supporting Real Time Resource Management  

1.   We have enhanced the Real-time CORBA support in the TAO Object Request Broker (ORB, Schmidt).

 

2.  We have investigated how the long range dependency (LRD) characteristics of Internet traffic --- existence of non-trivial correlation structure at multiple time scales --- can be judiciously exploited for better traffic control and measurement (Hou).  

 

back to top

Protocols and Infrastructure for Supporting Fault Tolerance

3.   We have laid an interception-based fault tolerant CORBA infrastructure, called Eternal system. The Eternal system intercepts the IIOP messages that would have been sent by the CORBA ORB and instead multicasts them to the replicas of an object using a reliable totally-ordered multicast protocol (Melliar-Smith and Moser).

 

4.   We have designed and implemented a pluggable fault toerant CORBA infrastructure that provides fault tolerance for CORBA applications by utilizing the pluggable protocols framework (PPF) (Melliar-Smith and Moser).

 

5.   We have developed the Totem Redundant Ring Protocol (Totem RRP) that supports three different styles of network replication: Active, Passive, and a hybrid Active-Passive network replication style (Melliar-Smith and Moser).

 

6.   We have developed the Aroma system, which provides transparent fault tolerance for Java applications that use Java Remote Method Invocation (RMI) (Melliar-Smith and Moser).  

back to top

Methodology and Protocols for Supporting Security

7.     We have developed a methodology for specification-based intrusion detection.  In previous work, we demonstrated the applicability of this method to detect attacks on privileged system calls and applications, such as relational database systems.  In the work on PERC, we have demonstrated the applicability of the method to detect attacks on protocols such as ARP (Address Resolution Protocol), DNS, and NFS (Levitt).

 

8.     We have devised a specification-based method to detect attacks on routers, in particular routers that utilize OSPF (Levitt). 

 

9.     We have defined JIGSAW, a language to specify multistage (scenario) attacks and HMAP, a system to detect misuse of the HTTP protocol (Levitt).

 

10.   We have developed IDIP, a technique to respond to attacks on a network by flooding filtering routers and firewalls with blocking rules (Levitt).

 

11.   We have developed a methodology to reason about specifications in support of intrusion detection.  Initially, for a Unix-based intrusion detection system, it is possible to show that the specifications for the intrusion detection system together with assumptions are sufficiently strong to detect unknown attacks (Levitt). 

 

back to top

Protocols for Supporting Scalable Real-Time Multicasts

12.  We have defined a multicast group communication engine and bridge that allows use of multiple group communication protocols concurrently (Melliar-Smoth and Moser).  

 

13. We have performed experimental measurements of the pluggable fault tolerant CORBA infrastructure and a reliable totally-ordered multicast group communication protocol.  Based on insights derived from these measurements, we have also developed a new replication strategy using semi-active replication (Melliar-Smith and Moser). 

 

14. We have developed mesh-based multicast protocols capable of supporting QoS and scaling to large numbers of nodes. This part of work is motivated by the observation that multicast routing trees are highly vulnerable in a wireless mobile network, because the failure of a single link or node breaks a tree (Garcia-Luna-Aceves).

 

15. We have laid a general temporal QoS framework for QoS-driven shared tree multicast routing and devised eligibility tests for member join/leave procedures (Hou).

 

16. We have proposed and investigated a novel protocol, called source-adaptive multi-layered multicast (SAMM), to control congestion caused by the real-time video multicast over the Internet (Suda).

back to top

Protocols for Supporting Mobile Wireless Environments

17. We have investigated participation incentives for proper routing in ad hoc networks. In particular, we have developed two general types of routing enhancements to detect and handle misbehaving nodes (Baker).

 

18. We have examined the queuing dynamics of nodes in ad hoc networks across a wide mobility spectrum in both on-demand and pro-active protocols and studied how to improve the performance of ad hoc networks through innovative packet scheduling. (Baker).

 

19. We have devised present a Minimum Spanning Tree (MST) based topology control algorithm, called Local Minimum Spanning Tree (LMST), for wireless multi-hop networks, and analytically proved several important properties of LMST (Hou).

 

20. We have studied the limitations of current contention-based channel access schemes and developed new channel access protocols capable of providing QoS in ad hoc networks with mobile nodes (Garcia-Luna-Aceves).

 

21. We have developed routing mechanisms that can provide QoS over multiple paths in ad hoc networks with mobile nodes. This part of our work addresses the issues associated with interconnecting ad hoc networks with the Internet, and managing battery life in untethered nodes (Garcia-Luna-Aceves)

 

22. We have applied aspects of dynamic and unpredictable environments in protocol designs in wireless and small device networks (Suda).

back to top

Bio-Networking Architectures

23. We have obtained deep understanding of how biological concepts and mechanisms can help network application technology and protocols, and shown that this new paradigm of applying biological concepts and mechanisms is feasible and promising (Suda).

 

24. We have addressed how discovery protocols may be adjusted to perform in a distributed and large-scale manner in peer-to-peer systems (Suda).  

back to top

Middleware for Supporting Application Programs with End-to-end QoS

25. We have developed a pluggable protocol framework for the TAO ORB (Schmidt).

 

26. We have added support for key QoS-enabled network protocols (including IntServ,         DiffServ, and SCTP) to TAO’s pluggable protocols framework (Schmidt).

 

27. We have laid the software architecture for the communication middleware to be developed (Hou).  

back to top

Tools for Supporting Network Simulation/Emulation

28. We have developed simulation techniques for very large application-level distributed protocols and begun implementation of Narses using network models of varying levels of details (Baker).

 

29. We have Completed the design and implementation of a component-based, compositional network simulation and emulation environment, JavaSim, and made the first formal release (after 8 pre-releases) in October 2001 (Hou).

 

back to top