7.6.1 Flow Controller Scheduling Policies

• Round Robin

  (DDS_RR_FLOW_CONTROLLER_SCHED_POLICY) Perform flow control in a round-robin (RR) fashion.

  Whenever tokens become available, the FlowController distributes the tokens uniformly across all of its (non-empty) destination queues. No destinations are prioritized. Instead, all destinations are treated equally and are serviced in a round-robin fashion.

• Earliest Deadline First

  (DDS_EDF_FLOW_CONTROLLER_SCHED_POLICY) Perform flow control in an earliest-deadline-first (EDF) fashion.

  A DDS sample's deadline is determined by the time it was written plus the latency budget of the DataWriter at the time of the write call (as specified in the DDS_LatencyBudgetQosPolicy). The relative priority of a flow controller's destination queue is determined by the earliest deadline across all DDS samples it contains.

  When tokens become available, the FlowController distributes tokens to the destination queues in order of their priority. In other words, the queue containing the DDS sample with the earliest deadline is serviced first. The number of tokens granted equals the number of tokens required to send the first DDS sample in the queue. Note that the priority of a queue may change as DDS samples are sent (i.e., removed from the queue). If a DDS sample must be sent to multiple destinations or two DDS samples have an equal deadline value, the corresponding destination queues are serviced in a round-robin fashion.

  With the default duration of 0 in the LatencyBudgetQosPolicy, using an EDF_FLOW_CONTROLLER_SCHED_POLICY FlowController preserves the order in which you call write() across the DataWriters associated with the FlowController.

  Since the LatencyBudgetQosPolicy is mutable, a DDS sample written second may contain an earlier deadline than the DDS sample written first if the DDS_LatencyBudgetQosPolicy’s duration is sufficiently decreased in between writing the two DDS samples. In that case, if the first DDS sample is not yet written (still in queue waiting for its turn), it inherits the priority corresponding to the (earlier) deadline from the second DDS sample.

  In other words, the priority of a destination queue is always determined by the earliest deadline among all DDS samples contained in the queue. This priority inheritance approach is required in order to both honor the updated duration and to adhere to the DataWriter in-order data delivery guarantee.

• Highest Priority First

  (DDS_HPF_FLOW_CONTROLLER_SCHED_POLICY) Perform flow control in an highest-priority-first (HPF) fashion.

  Note: Prioritized DDS samples are not supported when using the Ada API. Therefore, the Highest Priority First scheduling policy is not supported when using this API.

  The next destination queue to service is determined by the publication priority of the DataWriter, the channel of a multi-channel DataWriter, or individual DDS sample.

  The relative priority of a flow controller's destination queue is determined by the highest publication priority of all the DDS samples it contains.

  When tokens become available, the FlowController distributes tokens to the destination queues in order of their publication priority. The queue containing the DDS sample with the highest publication priority is serviced first. The number of tokens granted equals the number of tokens required to send the first DDS sample in the queue. Note that a queue’s priority may change as DDS samples are sent (i.e., as they are removed from the queue). If a DDS sample must be sent to multiple destinations or two DDS samples have the same publication priority, the corresponding destination queues are serviced in a round-robin fashion.

  This priority inheritance approach is required to both honor the designated publication priority and adhere to the DataWriter’s in-order data delivery guarantee.

  See also: 7.6.4 Prioritized DDS Samples.

© 2020 RTI