Learning Scheduling Algorithms for Data Processing Clusters


Hongzi Mao      Malte Schwarzkopf      Shaileshh Bojja Venkatakrishnan      Zili Meng     Mohammad Alizadeh     

Computer Science and Artificial Intelligence Laboratory
Massachusetts Institute of Technology


Abstract


Efficiently scheduling data processing jobs on distributed compute clusters requires complex algorithms. Current systems use simple, generalized heuristics and ignore workload characteristics, since developing and tuning a scheduling policy for each workload is infeasible. In this paper, we show that modern machine learning techniques can generate highly-efficient policies automatically. Decima uses reinforcement learning (RL) and neural networks to learn workload-specific scheduling algorithms without any human instruction beyond a high-level objective, such as minimizing average job completion time. However, off-the-shelf RL techniques cannot handle the complexity and scale of the scheduling problem. To build Decima, we had to develop new representations for jobs’ dependency graphs, design scalable RL models, and invent RL training methods for dealing with continuous stochastic job arrivals. Our prototype integration with Spark on a 25-node cluster shows that Decima improves average job completion time by at least 21% over hand-tuned scheduling heuristics, achieving up to 2× improvement during periods of high cluster load.


Paper


Learning Scheduling Algorithms for Data Processing Clusters
Hongzi Mao, Malte Schwarzkopf, Shaileshh Bojja Venkatakrishnan, Zili Meng, Mohammad Alizadeh
Proceedings of the 2019 ACM SIGCOMM Conference
[PDF]


Slides


SIGCOMM slides [PowerPoint] [PDF]
Extended slides [PowerPoint] [PDF]


Talk


[YouTube] [Download]


Poster


[Poster]


Code


[Code]


Press


MIT News, Tech Xplore, Computer Business Review, Science Blog, Network World and Notizie Scientifiche covered Decima.



Supporters


This project is supported by NSF, a Google Faculty Research Award, an AWS Machine Learning Research Award, a Cisco Research Center Award, an Alfred P. Sloan Research Fellowship and sponsors of MIT Data Systems and AI Lab.