Abstract: This work presents an adaptive profitable
discriminatory pricing mechanism for cloud computing based on
secure function decomposition, cryptographic commitments and
zero knowledge proof. Cloud computing is an emerging trend of
enterprise resource planning where a selling agent or service
provider (S) wants to allocate a set of computational resources
and related IT services optimally and fairly among many buying
agents or service consumers (B) within its capacity constraint.
Each service consumer discloses its demand plan for an IT
portfolio within its budget constraint and rank of preference. An
IT portfolio may include SaaS, PaaS, IaaS, CaaS, DaaS and
dSaaS. The basic objective of the service provider is to optimize
its expected revenue within target profit margin. It is basically a
problem of secure function evaluation where the concept of
decomposition of a function is considered. It is a constrained
nonlinear optimization problem; the search is governed by a set
of intelligent moves. The communication complexity of the
pricing mechanism depends on the time constraint of the
negotiating agents, their information state and the number of
negotiation issues; it also depends on number of negotiation
rounds and the complexity of IT portfolio. The computational
cost depends on the complexity of function decomposition. The
security and privacy of strategic data of the trading agents
provides business intelligence to the pricing mechanism. The
ultimate objective of the mechanism is to predict a profitable
discriminatory pricing plan for each consumer.
[Categories and Subject Descriptors] Pricing algorithm
[General Terms] Algorithmic mechanism
Keywords: Cloud computing, Nonlinear discriminatory
pricing mechanism, Computational intelligence.