The Benefit of Proactive Serving in User Delay Reduction

Joint work with Shaoquan Zhang from The Chinese University of Hong Kong, Longbo Huang from Tsinghua University, and Xin Liu from University of California, Davis.

In online service systems, the delay experienced by users from service request to service completion is one of the most critical performance metrics. To improve user delay experience, recent industrial practices suggest a modern system design mechanism: proactive serving, where the service system predicts future user requests and allocates its capacity to serve these upcoming requests proactively. This approach complements the conventional mechanism of capability boosting. In this paper, we propose queuing models for online service systems with proactive serving capability and characterize the user delay reduction by proactive serving. In particular, we show that proactive serving decreases average delay exponentially (as a function of the prediction window size) in the cases where service time follows light-tailed distributions. Furthermore, the exponential decrease in user delay is robust against prediction errors (in terms of miss detection and false alarm) and user demand fluctuation. Compared to the conventional mechanism of capability boosting, proactive serving is more effective in decreasing delay when the system is in the light-load regime. Our trace-driven evaluations demonstrate the practical power of proactive serving: for example, for the data trace of light-tailed Youtube videos, the average user delay decreases by 50% when the system predicts 60 seconds ahead. Our results provide, from a queuing-theoretical perspective, justifications for the practical application of proactive serving in online service systems.

In a sister paper, we apply the idea of “distribution shifting” to leverage prediction to improve delay performance in stochastic-optimization based scheduling algorithms.


  • S. Zhang, L. Huang, M. Chen, and X. Lin, “Proactive Serving Decreases User Delay Exponentially: The Light-tailed Service Time Case”, accepted for publication in IEEE/ACM Transactions on Networking. [PDF]

  • L. Huang, S. Zhang, M. Chen, and X. Liu, “When Backpressure Meets Predictive Scheduling”, IEEE/ACM Transactions on Networking, vol. 24, issue 4, Aug. 2016. [PDF] (Its conference version appeared on ACM MobiHoc 2014 as one of the final three Best Paper Award Candidates.)