Figure 8.2-1 outlines a detailed call flow for the AP-driven management of edge processing. In the previous sequence, the optional provisioning of an additional 5GMS AS instance occurs in response to an explicit call from the Application Client to an on-board EEC, whereas in this sequence it occurs, if needed, as part of and in response to Application Provider provisioning. The 5GMS AS instance may also be reprovisioned at any time during the lifetime of the 5GMS Provisioning Session in response to changing demand levels (not illustrated for brevity).
The Edge Computing Provisioning phase is a provisioning phase, that may be repeated several times (e.g. to extend edge processing coverage to new geographical areas or to increase the capacity of an already provisioned area). All steps in this phase are optional and performed on need basis. Steps 1-4 are identical to those described in clause 8.1 above.
The 5GMS Application Provider Provisioning phase is performed prior to the establishment of any related media streaming sessions by the 5GMS Application Provider. Subsequent updates to the provisioning session are possible. Steps 5-6 are identical to those described in clause 8.1 above.
The optional sub-flow to provision an additional 5GMS AS instance may be repeated multiple times on need basis to add new capacity, to increase existing capacity for edge processing or to reallocate underused edge processing resources to other tasks. The edge processing capacity is tailored for the specific 5GMS Application Provider based on the information in the Provisioning Session. Steps 7-12 are identical to steps 11-16 described in clause 8.1 above.
In step 7, based on the eligibility criteria in the edge resource template, the 5GMS AF shall determine whether the media streaming session is eligible to use edge resources.
After successful discovery, the actual streaming session may start in the 5GMS Session phase. Steps 13-15 are identical to steps 8-10 described in clause 8.1 above, and steps 16-21 are identical to steps 24-29.