43
Hardware Agnostic: Cassandra on Raspberry Pi Andy Cobley | Lecturer, University of Dundee, Scotland

C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

Embed Size (px)

DESCRIPTION

The raspberry Pi is a credit-card sized $25 ARM based linux box designed to teach children the basics of programming. The machine comes with a 700MHz ARM and 512Mb of memory and boots off a SD card, not much power for running the likes of a Cassandra cluster. This presentation will discuss the problems of getting Cassandra up and running on the Pi and will answer the all important question: Why on Earth would you want to do this!?

Citation preview

Page 1: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

Hardware Agnostic: Cassandra on Raspberry Pi

Andy Cobley | Lecturer, University of Dundee, Scotland

Page 2: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Cassandra is hardware agnostic *  So why not run it on a Raspberry Pi ? *  How hard can it be ? *  What can we do with it once it works?

Cassandra on Raspberry Pi

Page 3: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Andy Cobley *  School of Computing *  University of Dundee *  Twitter: @andycobley

Who Am I ?

Page 4: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Single chip Linux computer *  500 Meg ram *  Boots off an SD card *  Ethernet port *  (graphics and all you need for a general purpose computer)

Whats a Raspberry Pi ?

Page 5: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

Pi with pound coin

Page 6: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  And here’s the Cassandra cluster *

And, here’s one for real

* Power Permitting !

Page 7: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Cassandra is designed to be fast, fast at writing, fast at reading. *  This laptop with one instance of Cassandra will do 12,000 write

operations *  Raspberry Pi will do 200 !

The Bad News

Page 8: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Running a external USB drive is actually worse ! *  Probably be hardware feature

More bad news !

Page 9: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

Raspberry Pi Schematic

Page 10: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Oracle Java vs OpenJDK

And then there’s Java!

Page 11: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Raspbian is Debian for the PI *  Uses the Hard floating point accelerator *  Much faster than Debian *  Current Oracle JDK won’t run on it !

And Raspbian

Page 12: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  http://www.oracle.com/technetwork/java/embedded/downloads/javase/index.html

*  Java SE Embedded version 6 *  Cassandra might prefer 6 *  But *  https://blogs.oracle.com/henrik/entry/oracle_releases_jdk_for_linux *  Preview at: *  https://jdk8.java.net/fxarmpreview/

Oracle java

Page 13: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Actually not much difference in performance

Hard vs Soft Float

Page 14: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Cassandra uses compression for performance *  Started in version 1.0

2x-­‐4x  reduc+on  in  data  size  25-­‐35%  performance  improvement  on  reads  5-­‐10%  performance  improvement  on  writes  

The Problem with compression

Page 15: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Two types:

Google  Snappy  Compressor  (Faster  read/writes)  DeflateCompressor  (Java  zip,  slower  ,  beLer  compression)  

*  Snappy Compression not available on Pi

(requires  na+ve  methods,  so  someone  might  get  it  to  work!)  

Compression types

Page 16: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Startup script allocates memory *  Calculates based on number of processors *  Pi reports Zero processors ! *  Boom ! *  Now fixed

And the startup script

Page 17: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  In Cassandra-env.sh *  JVM_OPTS="$JVM_OPTS -

Djava.rmi.server.hostname=192.168.1.15” *  Or else nodetool will not work between nodes

JMX Config

Page 18: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  C* 1.22. added UseCondCardMark as a JVM Opt *  "for better lock handling especially on hotspot with multicore

processor” *  In cassandra-env.sh

#if  [  "$JVM_VERSION"  \>  "1.7"  ]  ;  then                                                                                                                                                #        JVM_OPTS="$JVM_OPTS  -­‐XX:+UseCondCardMark"                                                                                                                                #fi    

JVM OPT UseCondCardMark

Page 19: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley
Page 20: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  We’ve forgotten one thing *  The Pi cost £25 *  You can power 4 from USB hub (no need for a power supply on

each one) *  So:

The Good News !

Page 21: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

So, have a 64 node computer for £2000

University  of  Southhampton  

Page 22: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  32 node Beowolf cluster: *  Joshua Kiepert, Boise University

Or this

Page 23: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Adding nodes adds performance *  Adding nodes adds replicas of data *  BUT *  Make sure your ring is balanced, *  Pi’s don’t like to be unbalanced.

Adding nodes is good

Page 24: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Vnodes (in 1.2) would be very nice *  However at this point I haven’t got 1.2 on Pi running on a cluster

Vnodes

Page 25: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

Performance with 3/4 nodes

Page 26: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

Performance with 5/6 nodes

Page 27: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  ./stress -d 192.168.1.10,192.168.1.11,192.168.1.12 -o insert -I DeflateCompressor

*  Note: nodes to use *  You will get different performance if you insert to less nodes than

you have in your ring

Stress test commands

Page 28: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Adding a node (in the absence of Vnodes)

Must  seed  form  a  known  node  Use  a  program  to  calculate  new  keys    Bring  up  new  node  with  the  correct  key  in  cassandra.yaml  Use  node  tool  to  move  other  nodes  

Adding Nodes Procedure

Page 29: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Python code import  sys  if  (len(sys.argv)  >  1):        num  =  int(sys.argv[1])  else:        num  =  int(raw_input("How  many  nodes?  :"))  for  i  in  range(0,num):        print  'node  %d:  %d'  %  (i,  (i*(2**127)/num))  

Calculating keys

Page 30: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Use nodetool

sudo  ./nodetool  -­‐h  192.168.1.10  move  42535295865117307932921825928971026432  

*  And cleanup

./nodetool  -­‐h  192.168.1.10  cleanup  

Moving existing nodes

Page 31: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  On Debian, you can free memory from the graphics chip

Cd  /boot  sudo  cp  start.elf  start.elf.old  sudo  cp  arm224_start.elf  to  start.elf  reboot  

Getting more memory

Page 32: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Under Rasbian *  Run with a monitor plugged for the first time *  Set options for screen memory *  Perhaps disable boot to GUI

Getting more Memory

Page 33: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  I prefer static network addresses *  Edit /etc/network/interfaces iface  eth0  inet  sta+c                address  192.168.1.41                netmask  255.255.255.0                network  192.168.1.0                broadcast  192.168.1.255                gateway  192.168.1.254  

* 

Network address

Page 34: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Make a master SD card *  Copy it ! *  Make sure the master version has no data on it. *  Consider ”Puppet” (though I don’t use it)

Multiple nodes

Page 35: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  See https://github.com/acobley/CassandraStartup *  Put the file in /etc/init.d *  update-rc.d cassandra defaults

Starting as a service

Page 36: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  So for £200 we get an 8 node C* cluster *  It can be reconfigured, blown away, stress tested and generally

abused *  We can simulate data racks, data centers and I hope even long

network delays. *  Hopefully our upcoming MSc in Data Science will use these clusters

Pi is for teaching

Page 37: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  We know C* can be configured to be aware of:

Network  racks  Data  Centers  

*  We know we can have replicas are stored across these racks *  How can we play with this cheaply ?

C* is network aware

Page 38: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

Proposed teaching tool

10mbs  Hubb  

Noise  injec+on  

Switch  2  

Switch  1  

Pi  1  

Pi  2  

Pi  3  

Pi  1  

Pi  2  

Pi  3  

Page 39: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Cassandra wouldn’t run on a PI *  It does now. *  Running it on a Pi shook out some Cassandra bugs *  You can run it in a secure lab

Pi is discovery

Page 40: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Most important, this was pure Geeky Fun

Pi is for fun

Page 41: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Data Science: *  http://www.computing.dundee.ac.uk/study/postgrad/

degreedetails.asp?17

Obligatory Plug

Page 42: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

*  Raspberry Pi is cheap *  C* needs some work to run on it *  You can make clusters cheaply for experimentation *  It’s fun !

C* is Hardware Agnostic

Page 43: C* Summit 2013: Hardware Agnostic - Cassandra on Raspberry Pi by Andy Cobley

THANK YOU