模型并行

1. https://medium.com/@esaliya/model-paralelism-in-deep-learning-is-not-what-you-think-94d2f81e82ed

2. https://docs.microsoft.com/en-us/azure/machine-learning/concept-distributed-training

3. https://assets.amazon.science/ba/69/0a396bd3459294ad940a705ad7f5/herring-rethinking-the-parameter-server-at-scale-for-the-cloud.pdf