Swift on Diverse Infrastructures 
---------------------------------

Beagle
~~~~~~
Swift is available on Beagle as a software module.
In this recipe, we will walk you through a very simple example running Swift
on Beagle. The prerequisites required for running this example on Beagle is an access to Beagle and a project allocation. The example is a simplest form of a real application that involves reading from a disc file and writing onto another disc file. It runs the +/bin/cat+ Unix command to read an input file called +data.txt+ and write it to an output file in a folder +outdir+, called +f.nnn.out+.

image::figures/catsn.png[Swift Example]

The steps are as follows:

*step 1.* Load the Swift module on Beagle commandline as follows: +$ module load swift+

*step 2.* Create and change to a directory where your Swift related work (source, inputs, outputs) will
stay. (say, +$ mkdir swift-lab+, followed by, +$ cd swift-lab+)

*step 3.* copy the folder at +/home/ketan/labs/catsn+ to the above directory. (+$cp -r /home/ketan/catsn .+ followed by +$ cd catsn+).

*step 4.*  In the sites file: +sites.xml+ make the following two
changes:

 1. change the path of the +$$<workdirectory>$$+ element to your preferred location (this location is *required* to be on /lustre, say to +/lustre/beagle/<yourname>/swift-lab/swift.workdir+), and

  2. Change the project name to your project (for instance, +CI-CCR000013+) . The workdirectory will contain
  execution data related to each run, e.g. wrapper scripts, system information, inputs and outputs.

*step 5.* Run the example using following commandline (also found in run.sh):

----
$ swift -config cf -tc.file tc -sites.file sites.xml catsn.swift -n=1
----

You can further change the value of +-n+ to any arbitrary number to run that
many number of concurrent +cat+

*step 6.* Check the output in the generated +outdir+ directory (+ls outdir+)

Note: Running from sandbox node or requesting 1 hour walltime for upto 3 nodes
will get fast prioritized execution. Good for small tests
Swift supports a coasters based computing environment for Beagle. A detailed version of Swift documentation is maintained
http://www.ci.uchicago.edu/swift/docs/index.php[here].

//
//Intrepid-BG/P
//~~~~~~~~~~~~~
//Swift on Intrepid-BG/P
//

PADS
~~~~
Swift on PADS 
To execute your Swift script on the PADS cluster use this command: 

----
swift -tc.file tc -sites.file pbs.xml modis.swift
----
where the contents of a simple pbs.xml sites file could be:
[xml]
source~~~~
<config>
  <pool handle="pbs">
    <execution provider="pbs" url="none"/>
    <profile namespace="globus" key="queue">fast</profile>
    <profile namespace="globus" key="maxwalltime">00:05:00</profile>
    <profile namespace="karajan" key="initialScore">10000</profile>
    <profile namespace="karajan" key="jobThrottle">.10</profile>
    <filesystem provider="local"/>
    <workdirectory >/home/you/swiftwork</workdirectory>
  </pool>
</config>
source~~~~

OSG
~~~
This section describes how to get Swift running on the OSG Grid. We will use a
manual coaster setup to get Swift running on OSG.

.Coaster setup on OSG
The following figure shows an abstract scheme for the manual coasters setup on
OSG.

image::figures/coaster_setup.png[Coaster setup]

In the following steps, we will go through the process of manually setting 
//Swift on OSG 
//

Bionimbus
~~~~~~~~~
This section explains a step by step procedure on getting Swift running on the
Bionimbus cloud. We will use the _manual_ _coasters_ configuration on the
Bionimbus cloud.

**step1.** Connect to the gateway (ssh gatewayx.lac.uic.edu)

**step2.** Start a virtual machine (euca-run-instances -n 1 -t m1.small
emi-17EB1170)

**step3.** Start the coaster-service on gateway
+coaster-service -port 1984 -localport 35753 -nosec+

**step4.** Start the Swift-script from the gateway using normal Swift commandline

+swift -config cf -tc.file tc -sites.file sites.xml yourscript.swift -aparam=999+

.cf
----
wrapperlog.always.transfer=true
sitedir.keep=true
execution.retries=1
lazy.errors=true
status.mode=provider
use.provider.staging=true
provider.staging.pin.swiftfiles=false
foreach.max.threads=100
provenance.log=true
----

.tc
----
localhost modftdock /home/ketan/swift-labs/bionimbus-coaster-modftdock/app/modftdock.sh null null GLOBUS::maxwalltime="1:00:00"
----

(See below for a sample sites.xml for this run)

**step5.** Connect back to gateway from virtual machines using reverse ssh tunneling as follows:
 
.From the gateway prompt

+ssh -R *:5000:localhost:5000 root@10.101.8.50 sleep 999+

WHERE:
*=network interface, should remain the same on all cases

localhost=the gateway host, should remain the same

5000(LEFT OF localhost)=the port number on localhost to listen to **THIS WILL
vary depending upon which port you want to listen to

5000(RIGHT OF localhost)=the port on target host that you want to forward

root@10.101.8.50=the ip of the Virtual Machine on Bionimbus cloud, this will
vary based on what ip you get for your Virtual Machine instance

#On anywhere as long as provide the correct callback uri: here the
"http://140.221.9.110:42195" is the callback uri of previous ones

**step6.** Start the worker from the virtual machine
+worker.pl http://localhost:42195 tmp /tmp # where 42195 is the port where the
coaster service is listening to the workers+

.sites.xml for the above run

----
<config>
  <pool handle="localhost">
    <execution provider="coaster-persistent" url="http://localhost:1984" jobmanager="local:local"/>
    <profile namespace="globus" key="workerManager">passive</profile>

    <profile namespace="globus" key="workersPerNode">4</profile>
    <profile namespace="globus" key="maxTime">10000</profile>
    <profile namespace="globus" key="lowOverAllocation">100</profile>
    <profile namespace="globus" key="highOverAllocation">100</profile>
    <profile namespace="globus" key="slots">100</profile>
    <profile namespace="globus" key="nodeGranularity">1</profile>
    <profile namespace="globus" key="maxNodes">10</profile>
    <profile namespace="karajan" key="jobThrottle">25.00</profile>
    <profile namespace="karajan" key="initialScore">10000</profile>
    <profile namespace="swift" key="stagingMethod">proxy</profile>
    <filesystem provider="local"/>
    <workdirectory>/home/ketan/swift-labs/bionimbus-coaster-modftdock/swift.workdir</workdirectory>
  </pool>
</config>
----

//Magellan
//~~~~~~~~
//Swift on Magellan
//