System and method for arbitrating clients in a hierarchical real-time DRAM system

A method of arbitrating requests for access to a shared real-time resource received from a plurality of client devices, one of the clients being an aggregate client comprising a plurality of sub-clients, wherein the aggregate client arbitrates among the sub-clients to provide an access request to the shared resource, the method comprising:

• (a) imposing a minimum interval between successive access requests from the aggregate client, wherein the duration of the imposed minimum interval is substantially equal to DS/n, where DS is the duration of the shortest service deadline of the sub-clients and where the aggregate client aggregates n sub-clients;
• and
• (b) arbitrating access requests received from the plurality of clients.

A method of arbitrating requests for access to a shared real-time resource received from a plurality of client devices, one of the clients being an aggregate client comprising a plurality of sub-clients, wherein the aggregate client arbitrates among the sub-clients to provide an access request to the shared resource, the method comprising:

• (a) imposing a minimum interval between successive access requests from the aggregate client;
• and
• (b) arbitrating access requests received from the plurality of clients using one of rate monotonic scheduling and deadline monotonic scheduling, wherein a service deadline of the aggregate client, for purposes of the one of rate monotonic scheduling and deadline monotonic scheduling, is substantially equal to the duration of the imposed minimum interval.

A method of analyzing the schedulability of a real-time system that employs one of rate monotonic scheduling and deadline monotonic scheduling to determine access priority of a plurality of client devices to a shared resource, one of the clients being an aggregate client comprising a plurality of sub-clients, wherein the aggregate client arbitrates among the sub-clients to provide an access request to the shared resource, the method comprising:

• (a) imposing a minimum interval between successive access requests from the aggregate client, wherein the duration of the imposed minimum interval is substantially equal to DS/n where DS is the duration of the shortest service deadline of the sub-clients and where the aggregate client aggregates n sub-clients;
• and
• (b) performing a critical instant analysis of the system comprising the plurality of clients, wherein the imposed minimum interval is utilized as the service deadline of the aggregate client for purposes of the critical instant analysis.