A Strategy-proof Pricing Scheme for Multiple Resource Type Allocations

Abstract

Resource sharing on the Internet is becoming increasingly pervasive. Recently, there is growing interest in distributed systems such as peer-to-peer and grid, with efforts being directed towards resource allocation strategies that incentivize users to share resources. While combinatorial auctions can perform multiple resource type allocations, it is computationally a NPcomplete problem. Thus, allocation in large distributed resource sharing systems focuses mainly on a single resource type. We propose a strategy-proof, VCG-based resource pricing scheme for resource allocation in dynamic markets where users behave rationally in meeting their own interest. Our mechanism is designed to meet the needs of large distributed systems, delivering the following key properties: multiple resource type allocations, individual rationality, incentive compatibility for both buyers and sellers, budget balance and computational efficiency. Simulation evaluation of our prototype based on a centralized implementation demonstrates the viability of our approach, as compared to both traditional and combinatorial auctions.

The Impact of User Rationality in Federated Clouds

“…Accordingly, we have proposed a strategy-proof dynamic pricing scheme, where payments for both consumers and providers contain incentives for them to be truthful. Our proposed scheme achieved a higher percentage of allocated requests and a higher user welfare, when compared to fixed pricing [23]. Its economic properties, such as the allocation of consumer requests for multiple resource types, make it suitable in a federated cloud, where cloud resources from different providers are integrated for increased elasticity and reliability.…”

“…In a previous work, we proposed a strategy-proof pricing mechanism for allocating consumer requests for multiple resource types in federated clouds [15]. We based our payment scheme on Vickrey-Clarke-Groves, a well-known mechanism that incentivize rational users for being truthful, and used the mechanism design framework to prove its strategic and economic properties [23]. Other pricing schemes used in the allocation of shared resources that incentivize rational users for being truthful are based on combinatorial auctions [21], [28], or use the general equilibrium theory from economics to design market-based allocation schemes that provide mutual benefits to providers and consumers [12], [22].…”

“…Our study considers a federated cloud that consists of different Amazon EC2 regions. We analyze the spot pricing scheme used by Amazon EC2 to provide consumers with virtual machines called instances, and compare it with a strategy-proof dynamic pricing mechanism we propose to use in federated clouds [23]. Our main contributions are: i) a study of the rationality of cloud computing users, and ii) a model for determining EC2 workload using the spot pricing history.…”

Dynamic Pricing Scheme for Resource Allocation in Multi-Cloud Environment

“…In contrast, the dynamic pricing scheme is better than the fixed pricing scheme in terms of unpredictable demand [2,[8][9][10][11][12]. A strategy-proof dynamic scheme for resource allocation on federated clouds is proposed by [2].…”

A Distributed Market Framework for Large-Scale Resource Sharing

“…Specifically, we map a user buy request on the DHT lookup operation, and a user sell offer on the DHT store operation. For pricing, our framework makes use of a reverse auction-based mechanism that is able to allocate a request containing more than one resource types [21]. In addition to multiple resource type allocations, other properties achieved by the pricing mechanism are incentive compatibility and budget balance, at the expense of economic efficiency.…”