CASS mechanisms Sample Clauses

CASS mechanisms. In this section, we present the mechanisms that support CASS, including bidding constraints, QoS model, bid evaluation, winner determination, ask QoS generation and completion criteria. 4.3.1 Bidding constraints‌ In CASS, three bidding constraints are imposed: XOR bidding language, dynamic minimum increment and final deal-sealing round. 4.3.1.1 XOR bidding language 4.3.1.2 Dynamic minimum increment In each round, the auctioneer distributes ask QoS to the bidders. The bidders can choose to accept the ask QoS or revise their bids. And those that cannot afford the ask QoS will have to quit the auction. Ask QoS is used to coordinate the auction process. When generating the ask QoS, minimum increments (or decrements in the 3 Here “+” means the aggregation of QoS properties. case of negative properties), denoted by σ, can be imposed on the properties of the items. For example, “at least an increase of price by 5% than the previous bid” requires a bidder to bid a monetary price at least 5% higher than its previous bid in the last round. In the open SOC environment, some functionally-equivalent service providers may have huge different reserved capacities for the QoS while some others may have similar ones. In order for a combinatorial auction for service composition to be efficient, bidders with low reserved capacities should be filtered fast while bidders with high reserved capacities should be given more chances to propose better bids. In addition, the most notable approaches, which adopt Integer Programming (IP) and Mixed Integer Linear Programming (MILP) to solve the problem of QoS-aware service composition [5, 107, 108], are subject to computational uncertainty and scaling-up concerns. Specifically, there is no guarantee that the solution can be found in a reasonable amount of time when the number of bidders and items becomes larger. To address the above issues, we design a novel mechanism, named dynamic minimum increment (DMI). In general, at the early stage, higher minimum increment is adopted to generate ask QoS in order to efficiently filter the bidders with low capacities. By doing this, DMI efficiently decreases the number of bidders and bids at the early stage of the auction and thus reduces the complexity of the winner determination problem. And at the late stage, low minimum increment is adopted to distinguish the small difference among the remaining bidders’ reserved capacities, which guarantees the final outcome of the auctions (i.e. the quality of...