Summary of Cost-aware Dynamic Cloud Workflow Scheduling Using Self-attention and Evolutionary Reinforcement Learning, by Ya Shen et al.
Cost-Aware Dynamic Cloud Workflow Scheduling using Self-Attention and Evolutionary Reinforcement Learning
by Ya Shen, Gang Chen, Hui Ma, Mengjie Zhang
First submitted to arxiv on: 27 Sep 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Cost-aware Dynamic Multi-Workflow Scheduling (CDMWS) in the cloud tackles the challenge of assigning virtual machine (VM) instances to execute tasks in workflows while minimizing total costs. This includes penalties for violating Service Level Agreements (SLAs) and VM rental fees. The approach leverages deep neural networks and Reinforcement Learning (RL) methods to construct effective scheduling policies. A novel self-attention policy network, SPN-CWS, is introduced, which captures global information from all VMs. Additionally, an Evolution Strategy-based RL system is developed to train SPN-CWS reliably. The trained model can process multiple candidate VM instances simultaneously to identify the most suitable instance for each workflow task. Experimental results demonstrate that the proposed method outperforms several state-of-the-art algorithms on various benchmark CDMWS problems. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a new way to manage workflows in the cloud, called Cost-aware Dynamic Multi-Workflow Scheduling (CDMWS). This system tries to find the best virtual machines (VMs) to use for each task while keeping costs low. It’s like solving a puzzle with many moving parts. The researchers used special kinds of artificial intelligence called deep neural networks and Reinforcement Learning to create a smart decision-making tool. They also created a new type of network that can look at all the VMs together, not just one by one. This helps the system make better choices about which VM to use for each task. The results show that this approach works really well and is better than other methods used in similar situations. |
Keywords
» Artificial intelligence » Reinforcement learning » Self attention
