SparkDeepDocSome deep resources from apache spark, cloudera, my practice and so on. Most important is what i think.Streaming1: https://databricks.com/blog/2015/01/28/introducing-streaming-k-means-in-spark-1-2.html2: Flink 核心技术介绍 by Flink CommitterMachine learning1: K-D tree2: An overview of SGD3:trade-off of variance and bias