bnyeggen

(comment You will want just about everything in your hive/lib dir included in your Classpath)

(ns myproj.core
  (:import [org.apache.hadoop.hive.service HiveClient]
           [org.apache.thrift.transport TSocket]
           [org.apache.thrift.protocol TBinaryProtocol]))

(defn send-hive
  "Creates a new socket and Hive client connection, runs the query, pulls the result, and closes the connection.
  Eventually modify to split and parse according to schema of result.

bnyeggen

(comment Add [org.clojure/java.jdbc "0.1.1"] to project dependencies)

(ns myproject.core
  (:use [clojure.java.jdbc :only [with-connection, with-query-results]]))

(let [db-host "MyHost"
      db-port 10000
      db-name "default"]
  (def db {:classname "org.apache.hadoop.hive.jdbc.HiveDriver" ; must be in classpath
           :subname (str "//" db-host ":" db-port "/" db-name)

bnyeggen

from multiprocessing import Pool
from functools import partial

def _pickle_method(method):
	func_name = method.im_func.__name__
	obj = method.im_self
	cls = method.im_class
	if func_name.startswith('__') and not func_name.endswith('__'): #deal with mangled names
		cls_name = cls.__name__.lstrip('_')
		func_name = '_' + cls_name + func_name

bnyeggen

#redundancy is the max number of survivable failures, so eg 1 for RAID5
#mtbf_array is an array of either actual mean-time-between-failures, or a nested RAID array

# RAID([100]*7,2) #7 disk RAID 6
# RAID([RAID([100]*3,1),RAID([1000]*3,1)],0) # RAID 50, 2 arrays of 3
# RAID([100,100,50,50],1) #RAID 5 with varying reliabilities

from random import random

class RAID(object):