Auction mechanisms for allocation and pricing in Edge Computing
Time: 14:00 -- Location: LRI, 445
summary: Soon, billions of devices will be connected to the Internet, and most of them will need external resources to store and process their data. Edge Computing can benefit these devices by making these resources available at the network's edges to be closer to the user. However, these offered resources must be managed to be allocated efficiently. Auction mechanisms from Algorithmic Game Theory allow users to be modeled as buyers of Edge Computing resources being auctioned. To follow this approach, it is necessary to guarantee certain properties of an auction mechanism that are difficult to treat from a computational point of view. It is necessary to seek approximate allocation alternatives so that the auction is efficient while guaranteeing the maximization of resource usage. Therefore, the purpose is to develop approximate allocation mechanisms that are computationally efficient and have guarantees of maximizing the use of resources being offered in Edge Computing.