SparkInternals issues

[Doc] Fix a typo in 4-shuffleDetails.md

Fixed a typo in 4-shuffleDetails.md.

gitbook无法下载

https://www.gitbook.com/download/epub/book/yourtion/sparkinternals 这个链接点进去一直重定向到注册界面，登录了也无法下载。可以直接上传一下吗 ![image](https://user-images.githubusercontent.com/29473873/199284967-ec5c5d6f-caf3-485f-ae2f-5d1380e398dd.png)

MarkGao11520

Why the definition of dependencies is different from RDD paper?

From the paper [Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing](https://www.usenix.org/system/files/conference/nsdi12/nsdi12-final138.pdf) - narrow dependencies, where each partition of the parent RDD is used by at most one partition...

endersuu

Fix some typos

jaymmodi

Please add README.md to markdown/thai

1

This would be a great addition to the tahi books list at [Free-Programming-Books](https://github.com/EbookFoundation/free-programming-books/pull/6434). But the proposed link there makes it difficult to find the thai resource, and a link directly...

eshellman

Remove redundant word

Removed a redundant "blog" word.

GauthamBanasandra

还会继续更新吗？最后的两个章节？

1

RT

benchpress100

关于CogroupRDD的一点疑问以及依赖的一点问题

2

我看CogroupRDD的实现，没看懂narrowdependency或shuffledependency对cogrouprdd中partition的影响... 不知道如果a.cogroup(b) ， a分别是rangepartitioner和hashpartitioner的话，中间生成的cogrouprdd的分区数莫非和rdd a的一样多？因为cogroup这个算子不能指定numPartitons呀我看您在JobLogicalPlan章节中对dependency分了4类（或者说两打类), 而且看cogroupRDD的对于依赖的处理，似乎并没有这么复杂，完全无视了所谓的N:1 NarrowDependency。 > override def compute(s: Partition, context: TaskContext): Iterator[(K, Array[Iterable[_]])] = { > val sparkConf = SparkEnv.get.conf > val externalSorting = sparkConf.getBoolean("spark.shuffle.spill", true)...

pzz2011