Content delivery in wireless wide area networks
View Patent ↗Selecting an edge-server set in a wireless network comprises generating channel-quality measurements of wireless links between available nodes; determining a network topology state based on the measurements; calculating a performance metric for each candidate edge-server set; and selecting a candidate edge-server set based on the performance metric. An iterative process may be employed for selecting the best edge-server set. The iterative process may employ a trellis-exploration algorithm. A back-pressure routing algorithm may be used to calculate the performance metric. A server processes a request from a client residing on a wireless network by determining if another client on the wireless network has the requested object; determining if the requesting client can communicatively couple to the other client; and directing the request to the other client.
1. A method for selecting an edge-server set from a plurality of edge servers in a wireless network, comprising:
constructing a trellis having a number of states equal to a number of edge servers in the edge server set, wherein each state comprises a plurality of nodes, each node corresponding to one of a plurality of candidate edge servers;
and
employing a trellis-exploration algorithm to select the edge-server set, comprising providing interconnects between each node of a first state to each of a plurality of nodes in a next state, and for each node in a state, selecting a path corresponding to a best performance metric that connects to a node in a previous state, wherein each performance metric comprises a network topology state based on channel-quality measurements and wherein at least one of the plurality of candidate edge servers is selected based on which clients a candidate edge server can reach using peer-to-peer wireless links.
2. The method of claim 1 , wherein the network topology state comprises at least one of estimated bit rates, link-delay information, requests for retransmission, a wireless communication channel model, congestion, queue backlogs, and latency.
3. The method of claim 1 , wherein the network topology state comprises a statistical model of the network topology state comprising a time-average of at least one of server loads, user demands, queue backlogs, bit rates, channel quality, and geographical distributions of users.
4. The method of claim 1 , wherein the network topology state comprises at least one of a device network topology for each of a plurality of device types and a demand topology for each of a plurality of content types.
5. The method of claim 1 , wherein the network topology state is determined by at least one of active scanning and passive scanning.
6. The method of claim 1 , wherein the network topology state comprises data collected from client-side metrics managers.
7. The method of claim 1 , wherein the performance metric comprises monetary costs for employing at least one of the plurality of candidate edge servers.
8. A non-transitory machine-readable medium, comprising instructions for performing the method of claim 1 .
9. A method employing an iterative process for selecting an edge-server set, comprising:
constructing a trellis having a number of states at least equal to a number of edge servers in the edge server set, wherein each state comprises a plurality of nodes, each node corresponding to one of a plurality of candidate edge servers;
and
employing a trellis-exploration algorithm for selecting the edge-server set, comprising providing interconnects between each node of a first state to each of a plurality of nodes in a next state, and for each node in a state, selecting a path corresponding to a best performance metric that connects to a node in a previous state, wherein the best performance metric comprises a benefit minus a cost for serving clients and comprises a network topology state based on channel state information, and wherein at least one of the plurality of candidate edge servers is selected based on which clients a candidate edge server can reach using peer-to-peer wireless links.
10. The method of claim 9 , wherein the edge-server set is selected after a predetermined number of iterations or when a predetermined criterion is met.
11. The method of claim 9 , wherein selecting the edge-server set comprises at least one of appending and deleting at least one edge server from a previous edge-server set.
12. The method of claim 9 , wherein the performance metric is calculated using a backpressure routing algorithm.
13. A non-transitory machine-readable medium, comprising instructions for performing the method of claim 9 .
14. A method for selecting an edge server set, comprising:
constructing a trellis having a number of states at least equal to a number of edge servers in the edge server set, wherein each state comprises a plurality of nodes, each node corresponding to one of a plurality of candidate edge servers;
calculating a fitness function based on network performance improvements and costs corresponding to each candidate edge server; and
employing a trellis-exploration algorithm to identify a path through the trellis having optimal path metrics derived from the fitness function, comprising providing interconnects between each node of a first state to each of a plurality of nodes in a next state, and for each node in a state, selecting a path corresponding to a best path metric that connects to a node in a previous state, wherein the best path metric comprises the fitness function, and wherein at least one of the plurality of candidate edge servers is selected based on which clients a candidate edge server can reach using peer-to-peer wireless links.
15. The method of claim 14 , further comprising performing multiple iterations through the trellis to refine selecting the edge server set.
16. The method of claim 14 , wherein the network performance improvements comprise at least one of predicted performance improvements and measured performance improvements.
17. A non-transitory machine-readable medium, comprising instructions for peforming the method of claim 14 .