vinitshah24

#!/usr/bin/env bash
#
# Simulates mixed workload on HBase using YCSB
# Author: Ashrith (ashrith at cloudwick dot com)
# Date: Wed, 16 2014
#

#
# You may want to tweak these variables to change the workload's behavior
#

vinitshah24

import random

# Original array
array = [
    "category1/1", "category1/2", "category1/3", "category1/4", "category1/5", "category1/6",
    "category2/1", "category2/2", "category2/3", "category2/4",
    "category3/1", "category3/2"
]

# Separate elements by category

vinitshah24

import org.apache.hadoop.fs.{FileSystem, Path}

def removeFilesExceptFirstN(directory: String, prefix: String, n: Int): Unit = {
  val fs = FileSystem.get(sparkContext.hadoopConfiguration)
  val files = fs.globStatus(new Path(s"$directory/$prefix*")).map(_.getPath.toString).sorted
  files.drop(n).foreach({ file =>
    fs.delete(new Path(file), true)
  })
}

vinitshah24

import subprocess

# Define the BTEQ script
bteq_script = '''
.LOGON <Teradata_Hostname>/<Username>,<Password>;
SELECT InfoData FROM DBC.DBCInfo WHERE InfoKey = 'RELEASE';
.LOGOFF;
.EXIT;
'''