4 ms·
The person who wrote this article doesn't understand what the problem is, and how this paper doesn't address it at all. The paper has a better way to share one
by light_hue_1 3y ago
The person who wrote this article doesn't understand what the problem is, and how this paper doesn't address it at all.
The paper has a better way to share one GPU between small training jobs.
The real world problem is training huge models for a long time across many GPUs.
The problem the paper solves is one that is practically irrelevant aside from some niche cases.
- anotherthrway 3y agoThe problem to solve is GPU utilization rate. Training a “huge model” on n GPUs will have a cyclical pattern of idle GPUs and GPUs are expensive. This approach is likely scheduling GPU access across t training jobs to insure GPU is always in use. Maximizing GPU utility is a key requirement for training as a service. The biggest issue is that the failure mode of this approach (one fault hoses the entire pipeline of T jobs) is exactly opposing the goals of its most likely users: Training SaaS providers.