Routing (electronic design automation)

Algorithm

Two-level scheduling

Scheduling (production processes)

Operating system

Quality of service

Scalability

Computer science

Mathematical optimization

Distributed computing

Computer network

Reinforcement learning

Database

Rate-monotonic scheduling

Mathematics

Workload

Artificial intelligence

Job scheduler

Artificial neural network

Queue

Fair-share scheduling

Machine learning

Heuristics

Job shop scheduling

Efficiently scheduling data processing jobs on distributed compute clusters requires complex algorithms. Current systems, however, 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. Off-the-shelf RL techniques, however, 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 the average job completion time over hand-tuned scheduling heuristics by at least 21%, achieving up to 2x improvement during periods of high cluster load.

Learning scheduling algorithms for data processing clusters