1 个仓库
Awesome GitHub RepositoriesParallel Checkpoint Streams
Mechanisms for simultaneously writing large model states across multiple storage nodes.
Distinct from Model Checkpointing: Focuses on the parallel streaming nature of the write operation rather than the general persistence of model weights
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Parallel Checkpoint Streams. Refine with filters or upvote what's useful.
Awesome Parallel Checkpoint Streams GitHub Repositories
用 AI 发现最棒的仓库。我们将通过 AI 为您搜索最匹配的仓库。
- deepseek-ai/3fs
deepseek-ai/3FS
9,970在 GitHub 上查看3FS is a distributed file system and RDMA storage cluster designed for high-performance AI training and inference workloads. It functions as a strongly consistent storage layer that utilizes a disaggregated architecture to pool SSDs and memory resources across multiple nodes. The system provides specialized storage implementations including an AI training checkpoint store for parallel state preservation and a distributed key-value cache store for decoder layer vectors to optimize inference processing. It ensures data integrity through chain replication and apportioned query distribution. The
Writes large model states across multiple storage nodes simultaneously to minimize training downtime during synchronization.
C++
