Contracts Federated Cloud


We develop a new contracts-based resource sharing model for federated geo-distributed clouds that allows cloud service providers to establish contractual relationships with individual datacenters for defined time intervals. Based on the established contracts, individual cloud service providers employ a cost-aware job scheduling and provisioning algorithm that enables tasks to complete and meet their response time requirements.

What is the challenge?

  • How to decide Resource Sharing Contracts guaranteeing both Fairness and Efficiency?
  • How to quantify utility of the contracts for the Cloud Service Providers?
  • How to Schedule Jobs in a contracts-aware manner among the available Geo-distributed Data Centers?

How the System Works?

  • Auction-based Contracts Establishment (Designed based on McAfee Mechanism)
    • Truthfulness: No incentives to Cheat Makes the Market More Efficient and Fair
    • Budget Balance: Keeps the Resource Allocation process Sustainable
    • Naturally Finds the Equilibrium Between Supply and Demand
  • Fine-grained Utility Function Design considering:
    • Operating Cost
    • Payment from Users
    • Penalty for Violating SLAs
  • Resource Sharing Contracts
    • Relatively Long Contracts Avoid Frequent Migrations
    • Predetermined Allocations minimize Disruption of Normal Operations in the Data Centers
  • Cost-aware Scheduling
    • Optimize the Cost of Using Contracts



Jinlai Xu (徐锦来)
Jinlai Xu (徐锦来)

My research interests include Big Model Training/Inference/Finetune Framework, Serverless Computing, Distributed Systems, Fog/Edge and Cloud Computing, Stream Processing Optimization and Blockchain-based Techniques