kgorman

#!/usr/bin/env python

"""A simple python script template.

"""

import os
import sys
import argparse

kgorman

-- here is my stream
drop stream test_stream cascade;
create stream test_stream (speed int, ts int, tail char(10));

create continuous view v1 as select ts, tail, speed from test_stream;

create continuous view v2 as select tail, keyed_max(ts, speed) from test_stream group by tail;

 -- put some data into it
insert into test_stream values (100, 1, 'YY');

kgorman

brew install sbt

cd /tmp
wget http://apache.spd.co.il/kafka/0.8.0/kafka_2.8.0-0.8.0.tar.gz 
tar -zxvf kafka_2.8.0-0.8.0.tar.gz -C /usr/local/
cd /usr/local/kafka_2.8.0-0.8.0

sbt update
sbt package

kgorman

hour,avg_neg,avg_neutral,avg_pos
0,0.489045047812703,0.572904921392379,0.510954952187297
1,0.45979209657117,0.55928348584548,0.54020790342883
2,0.452954920632673,0.564895009665459,0.547045079367327
3,0.4886087921913,0.473499704794948,0.5113912078087
4,0.486929135727515,0.470766998533493,0.513070864272485
5,0.516430239594934,0.633499700744954,0.483569760405066
6,0.476850687564712,0.605231004647229,0.523149312435288
7,0.485249046873756,0.504506815590507,0.514750953126244
8,0.485469075837924,0.53258626109772,0.514530924162076

kgorman

hour,avg_neg,avg_neutral,avg_pos
0,0.489045047812703,0.572904921392379,0.510954952187297
1,0.45979209657117,0.55928348584548,0.54020790342883
2,0.452954920632673,0.564895009665459,0.547045079367327
3,0.4886087921913,0.473499704794948,0.5113912078087
4,0.486929135727515,0.470766998533493,0.513070864272485
5,0.516430239594934,0.633499700744954,0.483569760405066
6,0.476850687564712,0.605231004647229,0.523149312435288
7,0.485249046873756,0.504506815590507,0.514750953126244
8,0.485469075837924,0.53258626109772,0.514530924162076

kgorman

// mongodb 2.4.8


// fragmented
// initial setting of random rand of 5000
> use sbtest
switched to db sbtest
> db.stats()
{
	"db" : "sbtest",

kgorman

import sys

import os

sys.path.append("/usr/lib/spark-yarn/python")

os.environ["SPARK_HOME"]="/usr/lib/spark-yarn/"

from pyspark import SparkContext 

kgorman

# this script installs jupyter on Rackspace CBD
# 2015 kcg
# the trick here is to install python 3.3 with pip and not ruin the default python install
# then simply run jupyterhub
# this is designed to be run as a postinstall script when provisioning a cluster.

if [ $HOSTNAME == "GATEWAY-1" ]
  then

    # pre-requisites

kgorman

{
 "metadata": {
  "name": "",
  "signature": "sha256:a90804b68c0c02c405e62b261bc333673f0be01d2a735136d4799cd5e506f42d"
 },
 "nbformat": 3,
 "nbformat_minor": 0,
 "worksheets": [
  {
   "cells": [

kgorman

// a list of useful queries for profiler analysis.  Starting with the most basic.
// 2.4 compatible
//

// output explained:
//
{
 "ts" : ISODate("2012-09-14T16:34:00.010Z"),  			// date it occurred
	"op" : "query",						// the operation type
	"ns" : "game.players",					// the db and collection