LOECM

Citation Author(s):
Wenjia
Zhao
Submitted by:
Wenjia Zhao
Last updated:
Sat, 03/15/2025 - 10:59
DOI:
10.21227/tj5h-9p08
Data Format:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

Data centers are increasingly adopting renewable energy sources to mitigate environmental impact and reduce operational costs. However, effectively optimizing energy costs remains challenging due to unpredictable workloads and fluctuating renewable energy availability. This paper introduces LOECM, a Lyapunov-driven online scheduling algorithm designed to minimize energy cost without relying on future information. LOECM is formulated as an online stochastic optimization problem and leverages Lyapunov optimization techniques to balance immediate cost minimization against task queue stability. By exploiting the inherent structure of the optimization problem, we design an efficient solver with constant time and space complexity (O(1)). Additionally, we propose the LOECM scheduling policy to dynamically manage workloads and node activation based on a covering subset strategy. To evaluate the practicality and effectiveness of our approach, we implement a prototype system called LOECMHadoop, integrating the LOECM scheduler into Hadoop—a widely-used data processing platform—across a 10node cluster. Experimental results show that our power model has an average deviation of approximately 5%, and LOECM achieves over 20% electricity cost savings compared without our policy on the prototype system.

Instructions: 

The following .pyplot files record different aspects of the experimental data:

  • 2-3-4covering.pyplot: Records the runtime data under different Covering Subset configurations.
  • ac_en_cost.pyplot: Captures the active nodes, their energy consumption, and associated costs.
  • coveringsubset.pyplot: Logs the different processing capacities corresponding to different Covering Subset configurations.
  • ly-fifo-np-5fit-int.pyplot: Records the runtime performance of LOECMHadoop, FIFOHadoop, and NOptimalHadoop, allowing for comparative analysis of their efficiency and cost metrics.

 

These files provide valuable insights into system behavior, energy efficiency, and cost distribution across different scheduling strategies.

Comments

.

Submitted by ERKAN DURSUN on Tue, 03/18/2025 - 11:33

Dataset Files

    Files have not been uploaded for this dataset

    Documentation

    AttachmentSize
    File readme.txt785 bytes