You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Next »

1. Kepler - ITM vs. IMAS

Kepler installation for IMAS and ITM environments consists (mainly) of two sets of actors:

  • ITM: Kepler 2.4/2.5 based installation + custom ITM actors
  • IMAS: Kepler 2.4/2.5 based installation + custom IMAS actors

1.1. Setting up Kepler at Marconi Gateway

There are two, slightly different, ways of configuring Kepler at Marconi Gateway

ITMIMAS
Configuration of ITM Kepler at Marconi GatewayConfiguration of IMAS Kepler at Marconi Gateway

1.2. Working with Kepler at Marconi Gateway

Whenever you start Kepler at Marconi Gateway you can choose between one of two releases. Each of them is able to use underlaying data access layer. Below, you can take a brief look at most commonly used actors available for both platforms.

You can easily spot similarities in both sets. Each ITM based actor has corresponding IMAS based actors. Behavior of these actors is very similar. They simply operate on different data sets.

ITMIMAS
ITM actorIMAS actorDescription
ualinitUALInit

Initializes input pulse file, creates run work and provides IDS/CPO description for other actors

inputs:

    • user - name of the user for input data file (e.g. g2michal)
    • machine - name of the machine for input data file (e.g. test/jet)
    • shot - shot number
    • run - run number
    • runwork - temporary run number (place where output will be stored)

outputs:

    • error - description of error in case there are problems while accessing input data
    • all IDSes/CPOs requested by user (each IDS/CPO is specified as output port name)
ualcloseUALCloseCloses run work based on description passed via input IDS/CPO.
ualcollectorUALCollectorStores input IDSes/CPOs inside new run. This way, it is possible to copy data into different shot/run rather than inside run work.
ualslicecollectorUALSliceCollectorStore one slice from input IDS/CPO into different run. This way, it is possible to collect intermediate results during workflow execution.
ualmuxUALMux

Provides a method for putting data inside IDS/CPO

inputs:

    • inputIds/inputCpo - cpo we are going to modify
    • inTime - time index at which data are supposed to be updated
    • name of the field is specified as port name
    • new value of the field is passed as value sent to the port

outputs:

    • outputIds/outputCpo - modified IDS/CPO
    • outTime - actual time index (depend on approximation mode)
ualmuxparamUALMuxParam

Provides similar behavior to ualmux/UALMux, however, this actors has two additional ports:

    • fieldDescription - contains name of the filed that will be modified
    • fieldValue - contains new value of the field

Main difference between ualmux/ualmuxparam actors lays in it's ability to be used in a loop that modify different field inside IDS/CPO.

You can simply provide different field name for different loop's step.

ualdemuxUALDemuxAllows to read data from given IDS/CPO - name of the field is specified as output port.
ualpythonUALPython

Allows to run external Python process and pass input/output data between workflow and process itself. This actor is, most commonly, used for data visualization.

User can pass Python script directly to the actor.

setbreakpointSetBreakpointThis actor allows to enforce "debug" mode for IMAS/ITM based actors. It sets global parameter "ITM_DEBUG" to either true or false. In case true is a value of "ITM_DEBUG" all FC2K generated actors will start in debug mode.
cpooccurenceIDSOccurenceProvides a method to create duplicate of IDS/CPO with new occurrence number. This way, it is possible to store data before they get modified by user code.
cpoflushIDSFlushFlushes data from workflow. Data from memory cache are stored inside database.
cpodiscardIDSDiscardDiscards data inside workflow. Data will be re-read into memory cache.
cpocontentIDSContentDisplays IDS/CPO data.
  • No labels