MetaNet: Automated Dynamic Selection of Scheduling Policies in Cloud Environments
This work addresses cost-efficiency for cloud computing environments by enabling adaptive scheduling, though it is incremental as it builds on existing DNN-based schedulers.
The paper tackles the problem of high computational costs in deep neural network-based cloud task schedulers by proposing MetaNet, an automated system that dynamically selects scheduling policies online, resulting in improvements of up to 11% in execution costs, 43% in energy consumption, 8% in response time, and 13% in service level agreement violations compared to state-of-the-art methods.
Task scheduling is a well-studied problem in the context of optimizing the Quality of Service (QoS) of cloud computing environments. In order to sustain the rapid growth of computational demands, one of the most important QoS metrics for cloud schedulers is the execution cost. In this regard, several data-driven deep neural networks (DNNs) based schedulers have been proposed in recent years to allow scalable and efficient resource management in dynamic workload settings. However, optimal scheduling frequently relies on sophisticated DNNs with high computational needs implying higher execution costs. Further, even in non-stationary environments, sophisticated schedulers might not always be required and we could briefly rely on low-cost schedulers in the interest of cost-efficiency. Therefore, this work aims to solve the non-trivial meta problem of online dynamic selection of a scheduling policy using a surrogate model called MetaNet. Unlike traditional solutions with a fixed scheduling policy, MetaNet on-the-fly chooses a scheduler from a large set of DNN based methods to optimize task scheduling and execution costs in tandem. Compared to state-of-the-art DNN schedulers, this allows for improvement in execution costs, energy consumption, response time and service level agreement violations by up to 11, 43, 8 and 13 percent, respectively.