图存储的内存占用,以及split算子体现在何处?
- Dominant language
- C++
- Stars
- 2.9k
- Forks
- 553
- PR merge metrics
- No merged PRs in 30d
Description
想请教关于euler 2.0的几个问题
1. 目前的图结构加载起来后是以什么形式存在?是以邻接表的形式完全存在内存中吗,还是只存点和边的id在内存,具体的features单独存? 如果不是全部在内存中,是以什么形式,耗费的内存大概占图大小的多少比例呢?
2. 文档里面说的分布式自动添加split和merge算子的代码位置具体在哪里呢,我看IDSplit好像并没有被实际调用?这个op主要好处是dag分布式情况下优化起来更方便吗
3. 目前分布式下对于一些多跳采样的操作,看文档描述感觉仍然要等前一度的点返回了才能进行下一度的采样,感觉dag并行调度的优势似乎没有办法充分得到体现?
看了文档和一点点代码之后有些疑问,不知道是否愿意帮忙答疑一下,谢谢!
Contributor guide
No contributing guide indexed for this repository
Research direction
Start with the Euler 2.0 documentation and the IDSplit references mentioned in the issue. Trace where distributed split and merge operators are documented or invoked, then compare that with the descriptions of graph loading and multi-hop sampling. Done means the storage model, operator locations, and scheduling behavior are documented clearly.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- distributed-systems, documentation
- Issue type
- Documentation
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 20/100