The technical developments in computer networks in recent years have spawned the possibility of merging different services into a single Integrated Service Packet Network (ISPN). The types of service quality requires by each of the individual services in an ISPN often differ greatly. Thus, the packet scheduling algorithms used in such networks must be flexible enough to allocate the available link shares according to the service quality requirements of the different services. In this thesis we propose an efficient implementation of a link sharing algorithm for an Integrated Services Packet Switch (IMPS). The sharing algorithm guarantees to each customer a percentage of the link share no less than the portion of the bandwidth he owns. A customer can split its link share among different services whose link shares may in turn be split among child-services. The link sharing is hierarchical; each service gets its link from that of its parent service. Because of the complex nature of the link sharing the algorithm provides, the run time of its implementation can be large. However, in order to benefit from the fast links of networks like ATM, it is desirable to have reasonable short packet processing delays. The implementation presented here maps the hierarchical sharing into an equivalent single level sharing which resulted in a speed up of the packet processing executor by a factor of 10.
[1]
Hamid Ahmadi,et al.
A survey of modern high-performance switching techniques
,
1989,
IEEE J. Sel. Areas Commun..
[2]
John Nagle,et al.
On Packet Switches with Infinite Storage
,
1985,
IEEE Trans. Commun..
[3]
John B. Nagle,et al.
On Packet Switches with Infinite Storage
,
1987,
IEEE Trans. Commun..
[4]
Scott Shenker,et al.
Some observations on the dynamics of a congestion control algorithm
,
1990,
CCRV.
[5]
Scott Shenker,et al.
Supporting real-time applications in an Integrated Services Packet Network: architecture and mechanism
,
1992,
SIGCOMM '92.