1. Summary
  2. Files
  3. Support
  4. Report Spam
  5. Create account
  6. Log in

Utility Index

From cloudflu

Revision as of 06:45, 28 December 2010 by Alexey petrov (Talk | contribs)
Jump to: navigation, search

All the presented utilities have the following common set of options, namely :

  • --version show program's version number and exit.
    All the utilities have the same version number - library version. So, the easiest way to get known what is the version of the CloudFlu you are working with is to run a whatever CloudFlu utility with --version command-line option.
  • -h, --help show this help message and exit.
    Use this option to explore available command-line options, their default values and see corresponding usage example.
  • --debug turn on the debug information displaying.
  • --log-file=< file to write debug information >
  • --aws-access-key-id=< Amazon Key Id >
    Used for communication with Amazon Web Services. Check your Amazon account to obtain corresponding value.
  • --aws-secret-access-key=< Amazon Secret Key >
    Used for communication with Amazon Web Services. Check your Amazon account to obtain corresponding value.

Some of the command-line options are persistent, which means that it is possible to preset it globally from the corresponding preferences file. Such option will be highlighted by underlying. For example, the following common option are persistent one: --debug, --log-file, --aws-access-key-id and --aws-secret-access-key.

Contents

cloudflu-clean

This utility aims to clean all the possible cloud resources CloudFlu users could acquire. When it works it does not take care about whether this or that resource in use or not. So, use this utility with caution. The best practice is to call it each time at the end of your session.

It has no particular command-line options except common one, mentioned above.

Usage : cloudflu-clean

cloudflu-cluster-ls

This utility lists all the running clusters. As result user would see corresponding identifiers (something like r-0d6c3467). User could start as many clusters as he needs; by cloudflu-cluster-ls utility user always can refresh what clusters are already run.

It has no particular command-line options except common one, mentioned above.

Usage : cloudflu-cluster-ls

cloudflu-cluster-rm

This utility terminates appointed cluster by its identifier. The cluster identifier can be obtained as result of a_cluster_id=`cloudflu-cluster-start` && echo ${a_cluster_id} command or extracted from output of cloudflu-cluster-ls utility.

Extra options :

  • --cluster-id=< cluster identifier >
    (read from standard input, if not given)

Usage example :

  • cloudflu-cluster-rm --cluster-id=r-0d6c3467 # removes particular cluster
  • cloudflu-cluster-rm r-0d6c3467 # even shorter
  • cloudflu-cluster-ls | cloudflu-cluster-rm # removes all running clusters

cloudflu-cluster-start

Starts the a new cluster; returns corresponding cluster identifier as its output.

Extra options :

  • --instance-type=< EC2 instance type : 'c1.xlarge' or 'm1.large', for example >
    To see the difference among available instance types check http://aws.amazon.com/ec2/instance-types. Note : CloudFlu works only with 64 bit instance types.
  • --image-id=< Amazon EC2 AMI ID >
    Image contains as operation system as pre-installed software packages necessary to run OpenFOAM (R) use cases. Note : different images are responsible for representation of different OpenFOAM (R) versions; check your preferences file to choose the proper one.
  • --number-nodes=< number of cluster nodes >
    Actually it is most crucial cluster configuration parameter; user need to make some estimations before assign a value. Important : increasing number of cluster nodes does not directly lead to increasing performance; case partitioning need to be balanced between cluster node performance and the cost of inter-node data exchange. (We will be glad if you share with us your experience on this account to give us some direction and rules on how to estimate the number of cluster nodes in a formal way).

Usage example : cloudflu-cluster-start --instance-type='c1.xlarge' --image-id='ami-ecf50385' --number-nodes=8

cloudflu-config

Personal tools