Dinamica Console
Dinamica Console is the command-line interface for running Dinamica EGO model scripts without opening the graphical interface. It is well suited for batch processing, scheduled runs, and execution on machines without a display.
Syntax
DinamicaConsole {OPTIONS} <model script filename>
The <model script filename> argument is required and specifies the model script that will be executed. The special argument -- can be used to terminate flag options, forcing all following arguments to be treated as positional arguments.
Execution Options
-predefined-seed— Use a predefined value as the seed of the random number generators, instead of a value derived from the current time. Useful for producing reproducible runs.-granularity=[granularity]— Override the default granularity multiplier value used to break jobs into parallel tasks.-propagate-exceptions— Propagate top-level exceptions instead of only reporting them and exiting.-disable-submodel-expansions— Disable submodel expansions before running the script. See Submodel Expansion and Execution for details on what expansion does.-disable-copy-optimization— Disable data copy optimization during script execution. Only use this setting if you know what you are doing.-remove-redundant-carriers— Remove all unnecessary intermediate carrier functors before executing a model. See Submodel Expansion and Execution for details.-dont-run— Parse and validate the model script, but do not execute it. Useful together with-outputto only convert a script.
Workers and Memory
See Parallel Execution for how workers, granularity and accelerator devices are used.
-processors=[processors]— Override the number of logical processors (workers) that should be used. Pass 0 to automatically use the total number of processor cores instead.-disable-parallel-functors— Disable parallel processing of independent functors.-disable-parallel-steps— Disable parallel processing of independent loop steps.-disable-parallel-map-load— Disable loading maps in parallel.-memory-allocation-policy=[memory-allocation-policy]— Policy used to control memory allocation. Values:PREFER_MEMORY=0,BALANCED=1,PREFER_DISK=2,MEMORY_ONLY=3,AGGRESSIVE=4. See Memory Management for what each policy does;BALANCEDandAGGRESSIVEare generally recommended for performance, withAGGRESSIVEbeing the more performance-oriented andBALANCEDthe more conservative choice.-disable-native-expressions— Disable compilation of image expressions to native code. Expressions are evaluated using the default interpretation facility instead.
Accelerators
-accelerators-policy=[accelerators-policy]— Policy used to select accelerator (OpenCL) devices. Values:NO_DEVICES=0,ALL_DEVICES=1,USER_CONFIG=2.-list-accelerators— Show the available accelerator devices and exit, without running the script.
Logging
-log-level=[log-level]— Set the execution log level. Values:UNCONDITIONAL=0,ERROR=1,WARNING=2,RESULT=3,INFO=4,INFO2=5,DEBUG=6,DEBUG2=7. Note that this controls the volume of messages generated, not performance.-log-subsystem=[log-subsystem]— Set the execution log subsystem flags, restricting debugging information to specific subsystems. Only takes effect when-log-levelis set toDEBUGorDEBUG2. Values are bit flags that can be combined by summing them:GENERAL=1,OPENCL=2,SCRIPT_READER_WRITER=4,FUNCTOR_SCHEDULER=8,FUNCTOR_DEPENDENCE_GRAPH=16,FUNCTOR_EXECUTION=32,FILESYSTEM=64,MAP_IO_HANDLER=128,FUNCTOR_COPY_ANALYSIS=256,SUBMODEL=512,EXECUTION_BASICS=1024,GRAPHICAL_INTERFACE=2048,DATABASE=4096,TABLE_IO_HANDLER=8192,MAP_HANDLING=16384,TABLE_HANDLING=32768,SYSTEM_COMMUNICATION=65536. When the option is omitted, debugging information is written for all subsystems.-beep— Beep to indicate that the execution has completed.
During execution, the console only displays messages at the INFO2 level and below, regardless of the configured -log-level. Each worker also writes its own log file, named log_X.txt (where X is the worker number), which mirrors what the console shows: only messages at the INFO2 level and below, even if a DEBUG/DEBUG2 level is configured. When debug information is enabled, it is written separately to a per-worker file named debug_X.txt instead, following the same naming scheme; DEBUG and DEBUG2 messages never appear in log_X.txt or on the console. Both the log and debug files are created in the same folder as the model script being executed.
Metadata
Options for the metadata files written next to saved maps. See Map Metadata for the template keywords.
-metadata-profile=[metadata-profile]— Type of the profile used to write file metadata. Values:CUSTOM=0,CSR/UFMG_ptBR=1,CSR/UFMG_en=2.-map-metadata-template-filename=[filename]— Filename with the template used to write map metadata. Assumes the use of theCUSTOMprofile.-map-metadata-file-extension=[extension]— File extension used to write map metadata. Assumes the use of theCUSTOMprofile.
Model Output
-output=[output]— Write an output file representing the input model script, without needing to open the GUI to export it. The output format is chosen based on the file extension:.egofor EGO Script,.egomlfor EGOML Script,.dotfor a Graphviz representation of the model, and.egomlxfor a bundle script.-dont-run— Combine with-outputto only convert a script to another format without executing it.-verifier=[verifier]— Set the functor verifier used to validate the script before execution. Values:NonIntrusiveScriptVerifier,IntrusiveScriptVerifier.-write-semicolon-in-csv— Write CSV files using semicolons as value separators (and commas as decimal marks) instead of the English standard format.
Integration
-session-name=[session-name]— Create a session with the given name to share data between Dinamica EGO and an external R application via the “Calculate R Expression” functor.
Miscellaneous
-profile— Enable profiling the execution. When combined with an interactive terminal, this opens a text-based UI with tabs for the message log, worker status, accelerator usage, and resource (memory) tracking.-ask-to-start— Wait for a confirmation (press ENTER) before running the model.-ask-to-close— Wait for a confirmation (press ENTER) before closing the console window after execution finishes.-list-functors— Show the available functors and exit, without running the script.-version— Show version information and exit.+h,-help— Display the help menu.
Deprecated Options
The following options are accepted for backward compatibility but are ignored:
-scheduler=[scheduler]-disable-map-swapping
Examples
Run a model script using automatic worker detection and the default settings:
DinamicaConsole "C:\models\my_model.ego"
Run a model script with 4 workers, the Aggressive memory policy, and beep when finished:
DinamicaConsole -processors=4 -memory-allocation-policy=4 -beep "C:\models\my_model.ego"
Convert a model script to EGOML without running it:
DinamicaConsole -dont-run -output="C:\models\my_model.egoml" "C:\models\my_model.ego"
List the available accelerator devices:
DinamicaConsole -list-accelerators "C:\models\my_model.ego"
Run a model script with a reduced, more verbose log level for troubleshooting, restricted to the functor scheduler subsystem:
DinamicaConsole -log-level=6 -log-subsystem=8 "C:\models\my_model.ego"
Run a model script at the Debug level, logging multiple subsystems at once by summing their flags — here, GENERAL=1 + FUNCTOR_SCHEDULER=8 + MAP_IO_HANDLER=128 = 137:
DinamicaConsole -log-level=6 -log-subsystem=137 "C:\models\my_model.ego"