parallel-computing

parallel-computing.txt

### Docker ###

1) hello world:
   sudo docker run ubuntu:14.04 /bin/echo "hello, world!"

2) interactive:
   sudo docker run -t -i ubuntu:14.04 /bin/bash

3) daemonized hello world:
   sudo docker run -d ubuntu:14.04 /bin/bash -c "while true; do echo hello world; sleep 1; done"

   this will return a container id.
   a). sudo docker ps -l : list running containers. with name asigned.
   b). sudo docker logs container_name: show output of daemonized container.
   c). sudo docker stop container_name: stop daemonized container.

4) running a webapp in docker:
   sudo docker run -d -P training/webapp python app.py

   training/webapp is a pre-buildin image. will pulled from remote repository automatically if not found in local place.

   a) port binding: sudo docker port container_name [port_in_docker]
   b) show log continuously: sudo docker logs -f contianer_name
   c) look at container's process id: sudo docker top container_name
   d) inspect container: sudo docker inspect container_name
   e) stop and restart container: sudo docker stop/start container_name
   f) remove container: sudo docker rm container_name

5) docker images:
   a) search image from repository(default Docker Hub): sudo docker search centos

##############


### STORM ###
1) deploy a storm cluster on multi machines:
   a). setup a zookeeper cluster.
   b). start nimbus on master: bin/storm nimbus
   c). start supervisor on each worker machine: bin/storm supervisor
   d). start ui: bin/storm ui

#############


### ZOOKEEPER ###
1) clustered setup:
   a) config zookeeper: the default config file is at conf/zoo.cfg,
      modify zoo.cfg, and add myid file under data path:
      server.n = host:port1:port2
      content of file myid: n

      conf file sample:
        tickTime=2000
        dataDir=/var/lib/zookeeper
        clientPort=2181
        initLimit=5
        syncLimit=2
        server.1=zoo1:2888:3888
        server.2=zoo2:2888:3888
        server.3=zoo3:2888:3888
        
   b) start zookeeper on each machine: bin/zkServer.sh start

#################