Changes

From Genome Analysis Wiki
Jump to navigationJump to search
11,446 bytes removed ,  11:03, 29 October 2012
no edit summary
Line 33: Line 33:  
StarCluster creates a model configuration file in ~/.starcluster/config and you are instructed
 
StarCluster creates a model configuration file in ~/.starcluster/config and you are instructed
 
to edit this and set the correct values for the variables.
 
to edit this and set the correct values for the variables.
Here is an example of a config file that we used (with some details changed of course).
+
Here is a highly simplified example of a config file that should work.
 +
Please note there are many things you might want to choose, so craft the
 +
starcluster config file with care.
    
<code>
 
<code>
####################################
+
####################################
## StarCluster Configuration File ##
+
## StarCluster Configuration File ##
####################################
+
####################################
[global]
+
[global]
DEFAULT_TEMPLATE=xxx
+
DEFAULT_TEMPLATE=myexample
 
+
#############################################
+
#############################################
## AWS Credentials Settings
+
## AWS Credentials Settings
#############################################
+
#############################################
[aws info]
+
[aws info]
AWS_ACCESS_KEY_ID = AKImyexample8FHJJF2Q
+
AWS_ACCESS_KEY_ID = AKImyexample8FHJJF2Q
AWS_SECRET_ACCESS_KEY = fthis_was_my_example_secretMqkMIkJjFCIGf
+
AWS_SECRET_ACCESS_KEY = fthis_was_my_example_secretMqkMIkJjFCIGf
AWS_USER_ID=199998888709  
+
AWS_USER_ID=199998888709  
 
+
AWS_REGION_NAME = us-west-2                # Choose your own region
+
AWS_REGION_NAME = us-west-2                # Choose your own region
AWS_REGION_HOST = ec2.us-west-2.amazonaws.com
+
AWS_REGION_HOST = ec2.us-west-2.amazonaws.com
AWS_S3_HOST = s3-us-west-2.amazonaws.com
+
AWS_S3_HOST = s3-us-west-2.amazonaws.com
 
+
###########################
+
###########################
## EC2 Keypairs
+
## EC2 Keypairs
###########################
+
###########################
[key east1_starcluster]
+
[key west2_starcluster]
KEY_LOCATION = ~/.ssh/AWS/east1_starcluster_key.rsa
+
KEY_LOCATION = ~/.ssh/AWS/west2_starcluster_key.rsa   # Same region
 
+
[key west1_starcluster]
+
###########################################
KEY_LOCATION = ~/.ssh/AWS/west1_starcluster_key.rsa
+
## Define Cluster
 
+
  ##   starcluster start -c west2_starcluster  nameichose4cluster
[key west2_starcluster]
+
###########################################
KEY_LOCATION = ~/.ssh/AWS/west2_starcluster_key.rsa
+
[cluster myexample]
 
+
KEYNAME = west2_starcluster                # Name I chose
 
+
CLUSTER_SIZE = 8                            # Number of nodes
 
+
CLUSTER_SHELL = bash
# Configure the default cluster template to use when starting a cluster
+
# defaults to 'smallcluster' defined below. This template should be usable
+
# This is the 64 bit AMI from starcluster
# out-of-the-box provided you've configured your keypair correctly
+
NODE_IMAGE_ID = ami-c6bd50f6
DEFAULT_TEMPLATE=smallcluster
+
AVAILABILITY_ZONE = us-west-2a              # Region again!
# enable experimental features for this release
+
NODE_INSTANCE_TYPE = m1.large              # 8G memory should work for Pipeline
#ENABLE_EXPERIMENTAL=True
  −
# number of seconds to wait when polling instances (default: 30s)
  −
#REFRESH_INTERVAL=15
  −
# specify a web browser to launch when viewing spot history plots
  −
#WEB_BROWSER=chromium
  −
# split the config into multiple files
  −
#INCLUDE=~/.starcluster/aws, ~/.starcluster/keys, ~/.starcluster/vols
  −
 
  −
#############################################
  −
## AWS Credentials and Connection Settings ##
  −
#############################################
  −
[aws info]
  −
# This is the AWS credentials section (required).
  −
# These settings apply to all clusters
  −
# replace these with your AWS keys
  −
AWS_ACCESS_KEY_ID = #your_aws_access_key_id
  −
AWS_SECRET_ACCESS_KEY = #your_secret_access_key
  −
# replace this with your account number
  −
AWS_USER_ID= #your userid
  −
# Uncomment to specify a different Amazon AWS region (OPTIONAL)
  −
# (defaults to us-east-1 if not specified)
  −
# NOTE: AMIs have to be migrated!
  −
#AWS_REGION_NAME = eu-west-1
  −
#AWS_REGION_HOST = ec2.eu-west-1.amazonaws.com
  −
# Uncomment these settings when creating an instance-store (S3) AMI (OPTIONAL)
  −
#EC2_CERT = /path/to/your/cert-asdf0as9df092039asdfi02089.pem
  −
#EC2_PRIVATE_KEY = /path/to/your/pk-asdfasd890f200909.pem
  −
# Uncomment these settings to use a proxy host when connecting to AWS
  −
#AWS_PROXY = your.proxyhost.com
  −
#AWS_PROXY_PORT = 8080
  −
#AWS_PROXY_USER = yourproxyuser
  −
#AWS_PROXY_PASS = yourproxypass
  −
 
  −
###########################
  −
## Defining EC2 Keypairs ##
  −
###########################
  −
# Sections starting with "key" define your keypairs. See "starcluster createkey
  −
# --help" for instructions on how to create a new keypair. Section name should
  −
# match your key name e.g.:
  −
[key mykey]
  −
KEY_LOCATION=~/.ssh/mykey.rsa
  −
 
  −
# You can of course have multiple keypair sections
  −
# [key myotherkey]
  −
# KEY_LOCATION=~/.ssh/myotherkey.rsa
  −
 
  −
################################
  −
## Defining Cluster Templates ##
  −
################################
  −
# Sections starting with "cluster" represent a cluster template. These
  −
# "templates" are a collection of settings that define a single cluster
  −
# configuration and are used when creating and configuring a cluster. You can
  −
# change which template to use when creating your cluster using the -c option
  −
# to the start command:
  −
#
  −
#    $ starcluster start -c mediumcluster mycluster
  −
#
  −
# If a template is not specified then the template defined by DEFAULT_TEMPLATE
  −
# in the [global] section above is used. Below is the "default" template named
  −
# "smallcluster". You can rename it but dont forget to update the
  −
# DEFAULT_TEMPLATE setting in the [global] section above. See the next section
  −
# on defining multiple templates.
  −
 
  −
[cluster smallcluster]
  −
# change this to the name of one of the keypair sections defined above
  −
KEYNAME = mykey
  −
# number of ec2 instances to launch
  −
CLUSTER_SIZE = 2
  −
# create the following user on the cluster
  −
CLUSTER_USER = sgeadmin
  −
# optionally specify shell (defaults to bash)
  −
# (options: tcsh, zsh, csh, bash, ksh)
  −
CLUSTER_SHELL = bash
  −
# AMI to use for cluster nodes. These AMIs are for the us-east-1 region.
  −
# Use the 'listpublic' command to list StarCluster AMIs in other regions
  −
# The base i386 StarCluster AMI is ami-899d49e0
  −
# The base x86_64 StarCluster AMI is ami-999d49f0
  −
# The base HVM StarCluster AMI is ami-4583572c
  −
NODE_IMAGE_ID = ami-899d49e0
  −
# instance type for all cluster nodes
  −
# (options: cg1.4xlarge, c1.xlarge, m1.small, c1.medium, m2.xlarge, t1.micro, cc1.4xlarge, m1.medium, cc2.8xlarge, m1.large, m1.xlarge, m2.4xlarge, m2.2xlarge)
  −
NODE_INSTANCE_TYPE = m1.small
  −
# Uncomment to disable installing/configuring a queueing system on the
  −
# cluster (SGE)
  −
#DISABLE_QUEUE=True
  −
# Uncomment to specify a different instance type for the master node (OPTIONAL)
  −
# (defaults to NODE_INSTANCE_TYPE if not specified)
  −
#MASTER_INSTANCE_TYPE = m1.small
  −
# Uncomment to specify a separate AMI to use for the master node. (OPTIONAL)
  −
# (defaults to NODE_IMAGE_ID if not specified)
  −
#MASTER_IMAGE_ID = ami-899d49e0 (OPTIONAL)
  −
# availability zone to launch the cluster in (OPTIONAL)
  −
# (automatically determined based on volumes (if any) or
  −
# selected by Amazon if not specified)
  −
#AVAILABILITY_ZONE = us-east-1c
  −
# list of volumes to attach to the master node (OPTIONAL)
  −
# these volumes, if any, will be NFS shared to the worker nodes
  −
# see "Configuring EBS Volumes" below on how to define volume sections
  −
#VOLUMES = oceandata, biodata
  −
# list of plugins to load after StarCluster's default setup routines (OPTIONAL)
  −
# see "Configuring StarCluster Plugins" below on how to define plugin sections
  −
#PLUGINS = myplugin, myplugin2
  −
# list of permissions (or firewall rules) to apply to the cluster's security
  −
# group (OPTIONAL).
  −
#PERMISSIONS = ssh, http
  −
# Uncomment to always create a spot cluster when creating a new cluster from
  −
# this template. The following example will place a $0.50 bid for each spot
  −
# request.
  −
#SPOT_BID = 0.50
  −
 
  −
###########################################
  −
## Defining Additional Cluster Templates ##
  −
###########################################
  −
# You can also define multiple cluster templates. You can either supply all
  −
# configuration options as with smallcluster above, or create an
  −
# EXTENDS=<cluster_name> variable in the new cluster section to use all
  −
# settings from <cluster_name> as defaults. Below are example templates that
  −
# use the EXTENDS feature:
  −
 
  −
# [cluster mediumcluster]
  −
# Declares that this cluster uses smallcluster as defaults
  −
# EXTENDS=smallcluster
  −
# This section is the same as smallcluster except for the following settings:
  −
# KEYNAME=myotherkey
  −
# NODE_INSTANCE_TYPE = c1.xlarge
  −
# CLUSTER_SIZE=8
  −
# VOLUMES = biodata2
  −
 
  −
# [cluster largecluster]
  −
# Declares that this cluster uses mediumcluster as defaults
  −
# EXTENDS=mediumcluster
  −
# This section is the same as mediumcluster except for the following variables:
  −
# CLUSTER_SIZE=16
  −
 
  −
#############################
  −
## Configuring EBS Volumes ##
  −
#############################
  −
# StarCluster can attach one or more EBS volumes to the master and then
  −
# NFS_share these volumes to all of the worker nodes. A new [volume] section
  −
# must be created for each EBS volume you wish to use with StarCluser. The
  −
# section name is a tag for your volume. This tag is used in the VOLUMES
  −
# setting of a cluster template to declare that an EBS volume is to be mounted
  −
# and nfs shared on the cluster. (see the commented VOLUMES setting in the
  −
# example 'smallcluster' template above) Below are some examples of defining
  −
# and configuring EBS volumes to be used with StarCluster:
  −
 
  −
# Sections starting with "volume" define your EBS volumes
  −
# [volume biodata]
  −
# attach vol-c9999999 to /home on master node and NFS-shre to worker nodes
  −
# VOLUME_ID = vol-c999999
  −
# MOUNT_PATH = /home
  −
 
  −
# Same volume as above, but mounts to different location
  −
# [volume biodata2]
  −
# VOLUME_ID = vol-c999999
  −
# MOUNT_PATH = /opt/
  −
 
  −
# Another volume example
  −
# [volume oceandata]
  −
# VOLUME_ID = vol-d7777777
  −
# MOUNT_PATH = /mydata
  −
 
  −
# By default StarCluster will attempt first to mount the entire volume device,
  −
# failing that it will try the first partition. If you have more than one
  −
# partition you will need to set the PARTITION number, e.g.:
  −
# [volume oceandata]
  −
# VOLUME_ID = vol-d7777777
  −
# MOUNT_PATH = /mydata
  −
# PARTITION = 2
  −
 
  −
############################################
  −
## Configuring Security Group Permissions ##
  −
############################################
  −
# Sections starting with "permission" define security group rules to
  −
# automatically apply to newly created clusters. PROTOCOL in the following
  −
# examples can be can be: tcp, udp, or icmp. CIDR_IP defaults to 0.0.0.0/0 or
  −
# "open to the # world"
  −
 
  −
# open port 80 on the cluster to the world
  −
# [permission http]
  −
# PROTOCOL = tcp
  −
# FROM_PORT = 80
  −
# TO_PORT = 80
  −
 
  −
# open https on the cluster to the world
  −
# [permission https]
  −
# PROTOCOL = tcp
  −
# FROM_PORT = 443
  −
# TO_PORT = 443
  −
 
  −
# open port 80 on the cluster to an ip range using CIDR_IP
  −
# [permission http]
  −
# PROTOCOL = tcp
  −
# FROM_PORT = 80
  −
# TO_PORT = 80
  −
# CIDR_IP = 18.0.0.0/8
  −
 
  −
# restrict ssh access to a single ip address (<your_ip>)
  −
# [permission ssh]
  −
# PROTOCOL = tcp
  −
# FROM_PORT = 22
  −
# TO_PORT = 22
  −
# CIDR_IP = <your_ip>/32
  −
 
  −
 
  −
#####################################
  −
## Configuring StarCluster Plugins ##
  −
#####################################
  −
# Sections starting with "plugin" define a custom python class which perform
  −
# additional configurations to StarCluster's default routines. These plugins
  −
# can be assigned to a cluster template to customize the setup procedure when
  −
# starting a cluster from this template (see the commented PLUGINS setting in
  −
# the 'smallcluster' template above). Below is an example of defining a user
  −
# plugin called 'myplugin':
  −
 
  −
# [plugin myplugin]
  −
# NOTE: myplugin module must either live in ~/.starcluster/plugins or be
  −
# on your PYTHONPATH
  −
# SETUP_CLASS = myplugin.SetupClass
  −
# extra settings are passed as __init__ arguments to your plugin:
  −
# SOME_PARAM_FOR_MY_PLUGIN = 1
  −
# SOME_OTHER_PARAM = 2
  −
 
  −
######################
  −
## Built-in Plugins ##
  −
######################
  −
# The following plugins ship with StarCluster and should work out-of-the-box.
  −
# Uncomment as needed. Don't forget to update your PLUGINS list!
  −
# See http://web.mit.edu/star/cluster/docs/latest/plugins for plugin details.
  −
#
  −
# Use this plugin to install one or more packages on all nodes
  −
# [plugin pkginstaller]
  −
# SETUP_CLASS = starcluster.plugins.pkginstaller.PackageInstaller
  −
# # list of apt-get installable packages
  −
# PACKAGES = mongodb, python-pymongo
  −
#
  −
# Use this plugin to create one or more cluster users and download all user ssh
  −
# keys to $HOME/.starcluster/user_keys/<cluster>-<region>.tar.gz
  −
# [plugin createusers]
  −
# SETUP_CLASS = starcluster.plugins.users.CreateUsers
  −
# NUM_USERS = 30
  −
# # you can also comment out NUM_USERS and specify exact usernames, e.g.
  −
# # usernames = linus, tux, larry
  −
# DOWNLOAD_KEYS = True
  −
#
  −
# Use this plugin to configure the Condor queueing system
  −
# [plugin condor]
  −
# SETUP_CLASS = starcluster.plugins.condor.CondorPlugin
  −
#
  −
# The SGE plugin is enabled by default and not strictly required. Only use this
  −
# if you want to tweak advanced settings in which case you should also set
  −
# DISABLE_QUEUE=TRUE in your cluster template. See the plugin doc for more
  −
# details.
  −
# [plugin sge]
  −
# SETUP_CLASS = starcluster.plugins.sge.SGEPlugin
  −
# MASTER_IS_EXEC_HOST = False
  −
#
  −
# The IPCluster plugin configures a parallel IPython cluster with optional
  −
# web notebook support. This allows you to run Python code in parallel with low
  −
# latency message passing via ZeroMQ.
  −
# [plugin ipcluster]
  −
# SETUP_CLASS = starcluster.plugins.ipcluster.IPCluster
  −
# ENABLE_NOTEBOOK = True
  −
# #set a password for the notebook for increased security
  −
# NOTEBOOK_PASSWD = a-secret-password
  −
#
  −
# Use this plugin to create a cluster SSH "dashboard" using tmux. The plugin
  −
# creates a tmux session on the master node that automatically connects to all
  −
# the worker nodes over SSH. Attaching to the session shows a separate window
  −
# for each node and each window is logged into the node via SSH.
  −
# [plugin tmux]
  −
# SETUP_CLASS = starcluster.plugins.tmux.TmuxControlCenter
  −
#
  −
# Use this plugin to change the default MPI implementation on the
  −
# cluster from OpenMPI to MPICH2.
  −
# [plugin mpich2]
  −
# SETUP_CLASS = starcluster.plugins.mpich2.MPICH2Setup
  −
#
  −
# Configure a hadoop cluster. (includes dumbo setup)
  −
# [plugin hadoop]
  −
# SETUP_CLASS = starcluster.plugins.hadoop.Hadoop
  −
#
  −
# Configure a distributed MySQL Cluster
  −
# [plugin mysqlcluster]
  −
# SETUP_CLASS = starcluster.plugins.mysql.MysqlCluster
  −
# NUM_REPLICAS = 2
  −
# DATA_MEMORY = 80M
  −
# INDEX_MEMORY = 18M
  −
# DUMP_FILE = test.sql
  −
# DUMP_INTERVAL = 60
  −
# DEDICATED_QUERY = True
  −
# NUM_DATA_NODES = 2
  −
#
  −
# Install and setup an Xvfb server on each cluster node
  −
# [plugin xvfb]
  −
# SETUP_CLASS = starcluster.plugins.xvfb.XvfbSetup
      
</code>
 
</code>
283

edits

Navigation menu