Abstract
Migrating Sockets is the protocol processing component of an end system architecture designed for networking with QoS guarantees. The architecture provides: (1) adaptive rate-controlled scheduling of protocol threads in Migrating Sockets, (2) rate-based flow control for reserved rate connections in future integrated services networks, and (3) a constant overhead active demultiplexing mechanism. Migrating Sockets achieves its efficiency by allowing user applications to manage a network endpoint with minimal system intervention, providing user level protocols read-only access to routing information in a well-known shared memory region, and integrating efficient kernel level support we previously built. It is backward compatible with Unix semantics and Berkeley sockets, and has been used to implement Internet protocols such as TCP, UDP and IP (including IP multicast). We also show that active demultiplexing supported by Migrating Sockets can be transparently enabled in wide-area TCP/IP internetworking (although it is not restricted to TCP/IP). We have an implementation of Migrating Sockets in Solaris 2.5. We discuss our implementation experience, and present performance results of our system running on the Ultra-1, SPARC 10 and SPARC 20 architectures.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.