the fastest distributed algorithm for building the timeseries index
following are a brief installation guide. I will provide my papers in another url the next time with algorithm description and replace this with a detailed installation instructions. My main idea is included in sparkindex/sparkIndex.scala。
1.copy spark libs and hbase libs into libs
2.intellij with scala plugin
3.scala 10.4+
4.Java1.7 +
5.package the project into a jar(named SparkIsax.jar)
because of the size limit on the github, i didn't push my libs and jars.
spark-submit --master spark://master:7077 SparkIsax.jar tableName inputPath MaximumDepth