Code coupling in Kraken

The code coupling in Kraken is based on a "hub and spoke" approach where each solver module only needs to communicate with the central multi-physics driver Cerberus. The Cerberus multi-physics driver is responsible for starting up each solver module, exchanging field data with them and controlling the execution order of the different modules.

The individual solver modules should implement basic signalling capabilities so that they can accept signals from Cerberus as well as send messages back using the signalling syntax described below.

Basics of signalling

When the solver is executed from command line by Cerberus when running a Python script such as

# --- Initialize solvers

my_input = CodeInput("path_to_my_input")
my_solver = Solver("my_solver_name", path_to_my_solver, ["--port",])
my_solver.input = my_input
my_solver.initialize()

Cerberus will provide the port the solver should connect to as a pair of command line arguments "--port" <port_as_integer>. The solver should then read its own input, do whatever processing is required and open a socket on the indicated port for communication with Cerberus.

After the port has been opened, Cerberus expects the following greeting process:

Direction	Name	Size and type	Content
Solver->Cerberus	Lgreet	1*integer	Length of greeting string in char excluding terminating character
Solver->Cerberus		Lgreet*char	A freely chosen greeting string (will be read and can be printed out by Cerberus).
Cerberus->Solver	Lgreet	1*integer	Length of greeting string in char excluding terminating character
Cerberus->Solver		Lgreet*char	Some greeting string sent by Cerberus.

After the greeting and its own basic initialization, the solver should wait for further signals from Cerberus.

After each order received from Cerberus, the solver should fulfill the order and wait for further signals from Cerberus.

All numerical values will be passed as double precision integers (8 bytes) or double precision floats (8 bytes).

Basic "function calls"

When the solvers have completed their previous order and are waiting for the next signal from Cerberus, Cerberus will initiate the communication with a single integer indicating which "function call" should be processed.

The function calls and the integers representing them in the communication syntax are listed in a table below. When coupling a solver to Cerberus, they are typically listed in some centrally located header (or some such) files for C and Fortran. These can be also checked from cerberus.commons.KrakenSignals.

Function name	Integer value
terminate	0
returnCurrentTime	1
setCurrentTime	2
suggestNextTime	3
advanceToTime	4
solveCurrentTime	5
giveInputFieldNames	6
giveOutputFieldNames	7
giveFieldTemplate	8
giveMeshData	9
giveIndexingArray	10
giveFieldData	11
takeFieldData	12
giveOutputVariableDefs	13
giveVariableData	14
giveInputVariableDefs	15
takeVariableData	16
writeRestart	17
readRestart	18
returnCurrentTimeInterval	19
setCurrentTimeInterval	20
suggestNextTimeInterval	21
advanceToTimeInterval	22
correctPreviousStep	23

The following sections will go through the signalling for each of the function calls

terminate

When Cerberus sends the signal for a solver to terminate. The solver simply exits gracefully. Final results can be collected, processed and printed out. Memory should be freed, and the process should be terminated in the end.

No further signalling takes place after the C->S signal to terminate.

giveInputFieldNames

The solver should communicate the names of the fields it is expecting to receive from Cerberus. The communication syntax is as follows:

Direction	Name	Size and type	Content
S->C	NF	1*integer	Number of fields to be provided
The following repeats NF times:
S->C	Lname	1*integer	Length of field name in char excluding terminating character
S->C		Lname*char	Name of the field

giveOutputFieldNames

The solver should communicate the names of the fields it is expecting to provide to Cerberus. The communication syntax is as follows:

Direction	Name	Size and type	Content
S->C	NF	1*integer	Number of fields to be provided
The following repeats NF times:
S->C	Lname	1*integer	Length of field name in char excluding terminating character
S->C		Lname*char	Name of the field

giveFieldTemplate

The solver should provide basic information on the field requested by Cerberus.

The communication syntax is as follows:

Direction	Name	Size and type	Content
C->S	LnameF	1*integer	Length of name of requested field (in char excluding terminating character).
C->S		LnameF*char	Name of the requested field
S->C		1*integer	Data type (FLOAT = 1, INT = 2, STRING = 3)
S->C		5*integer	Unit of the field in an OpenFOAMish array (see Field unit array)
S->C		1*float	Multiplier a used to convert from the base unit type specified by the unit array to the actual unit used by the solver (a*x + b).
S->C		1*float	Offset b used to convert from the base unit type specified by the unit array to the actual unit used by the solver (a*x + b).
S->C	LnameM	1*integer	Length of mesh name in char excluding terminating character
S->C		LnameM*char	Name of the mesh
S->C		1*integer	Indexing flag: 0 if solver uses the default indexing of the mesh. 1 if solver uses some other indexing: Indexing array
[S->C]	LnameI	1*integer	Only if indexing flag was given as 1 (use custom indexing). Length of indexing array name in char excluding terminating character.
[S->C]		LnameI*char	Only if indexing flag was given as 1 (use custom indexing). Name of the indexing array.

giveMeshData

The solver should provide information on the mesh requested by Cerberus.

The communication syntax is as follows:

Direction	Name	Size and type	Content
C->S	Lname	1*integer	Length of name of requested mesh (in char excluding terminating character).
C->S		Lname*char	Name of the requested mesh
S->C		1*integer	Type of the mesh (see Mesh types).
S->C		<depends on mesh>	Mesh data, see subsection Mesh types

giveIndexingArray

The solver should provide information on the indexing array requested by Cerberus.

The communication syntax is as follows:

Direction	Name	Size and type	Content
C->S	Lname	1*integer	Length of name of requested indexing array (in char excluding terminating character).
C->S		Lname*char	Name of the requested indexing array
S->C	NV	1*integer	Number of values to be passed.
S->C		NV*value	Indices: Indexing array

giveFieldData

The solver should provide the data for the field requested by Cerberus.

The communication syntax is as follows:

Direction	Name	Size and type	Content
C->S	Lname	1*integer	Length of name of field to be passed (in char excluding terminating character).
C->S		Lname*char	Name of the field to be passed
S->C	NV	1*integer	Number of values to be passed.
S->C		NV*value	The values for the field (float, integer or strings).

takeFieldData

The solver should receive the data for the field indicated by Cerberus.

The communication syntax is as follows:

Direction	Name	Size and type	Content
C->S	Lname	1*integer	Length of name of field to be passed (in char excluding terminating character).
C->S		Lname*char	Name of the field to be passed
C->S	NV	1*integer	Number of values to be passed.
C->S		NV*value	The values for the field (float, integer or strings).

giveOutputVariableDefs

The solver should communicate names of the variables it is expecting to provide to Cerberus. The communication syntax is as follows:

Direction	Name	Size and type	Content
S->C	NV	1*integer	Number of variable definitions to be provided
The following repeats NV times:
S->C	Lname	1*integer	Length of name of variable to be passed (in char excluding terminating character).
S->C		Lname*char	Name of the variable to be passed
S->C		1*integer	Data type (FLOAT = 1, INT = 2, STRING = 3)
S->C		1*integer	Number of values to be passed.

giveVariableData

The solver should provide the variable requested by Cerberus.

The communication syntax is as follows:

Direction	Name	Size and type	Content
C->S	Lname	1*integer	Length of name of variable to be passed (in char excluding terminating character).
C->S		Lname*char	Name of the variable to be passed
S->C	NV	1*integer	Number of values to be passed.
S->C		NV*value	Variable values (float, integer or strings).
S->C		NV*float	Only if statistics flag was given as 1. Relative errors.

giveInputVariableDefs

The solver should communicate names of the variables it is expecting to accept from Cerberus. The communication syntax is as follows:

Direction	Name	Size and type	Content
S->C	NV	1*integer	Number of variable definitions to be provided
The following repeats NV times:
S->C	Lname	1*integer	Length of the name of the variable to be passed (in char excluding terminating character).
S->C		Lname*char	Name of the variable to be passed
S->C		1*integer	Data type (FLOAT = 1, INT = 2, STRING = 3)
S->C		1*integer	Number of values to be passed.

takeVariableData

The solver should receive the data for the variable indicated by Cerberus.

The communication syntax is as follows:

Direction	Name	Size and type	Content
C->S	Lname	1*integer	Length of the name of the variable to be passed (in char excluding terminating character).
C->S		Lname*char	Name of the variable to be passed
C->S	NV	1*integer	Number of values to be passed.
C->S		NV*value	Variable values (float, integer or strings).

returnCurrentTime

The solver should return the time corresponding to the current time point as a float in seconds. This can be used to check that all solvers proceed at the same time points in burnup calculations.

Direction	Name	Size and type	Content
S->C	t1	1*float	Current time point of the solver (seconds).

setCurrentTime

This method is usually called in the beginning of burnup calculations to set the starting time of the simulation. The solver should store the received time (a float in seconds) as the current time point and conduct any additional processing required.

Direction	Name	Size and type	Content
C->S	t1	1*float	Time point to be set (seconds).

suggestNextTime

The solver should return a preferred length for the next time point (end of next timestep) as a float in seconds.

Direction	Name	Size and type	Content
S->C	t1	1*float	Suggested next time point (seconds).

advanceToTime

The solver should advance to the next time point (time step), with the given value in seconds. This process may involve storing final results for the previous time point (time step), time integration of some data forwards to the next time point (e.g. burnup accumulation) etc.

Direction	Name	Size and type	Content
C->S	t1	1*float	Next time point (seconds).

returnCurrentTimeInterval

The solver should provide information on the time interval it is currently solving to Cerberus (used in transient calculations):

Direction	Name	Size and type	Content
S->C	t0	1*float	Beginning time of current interval (seconds).
S->C	t1	1*float	End time of current interval (seconds).

setCurrentTimeInterval

Cerberus provides the solver information on the time interval to be solved and the solver should adjust itself accordingly (used in transient calculations):

Direction	Name	Size and type	Content
C->S	t0	1*float	Beginning time of current interval (seconds).
C->S	t1	1*float	End time of current interval (seconds).

suggestNextTimeInterval

The solver should provide its suggestion for the next time interval to be solved (used in transient calculations):

Direction	Name	Size and type	Content
S->C	t0	1*float	Suggested beginning time of next time interval (seconds).
S->C	t1	1*float	Suggested end time of next time interval (seconds).

The beginning time of the next time interval should be equal to the end time of the current time interval.

advanceToTimeInterval

Cerberus provides the solver information on the next time interval to be solved. The solver should prepare itself for solving the new time interval next (e.g. by setting the end-of-interval data of the previous time-interval to be used as beginning-of-interval data for the upcoming one). Used in transient calculations.

Direction	Name	Size and type	Content
C->S	t0	1*float	Beginning time of the next time interval (seconds).
C->S	t1	1*float	End time of the next time interval (seconds).

The beginning time of the next interval is more or less guaranteed to be the end time of the current interval.

solveCurrentTime

The solver should execute the solution for the current time point (time step). This can include some initialization and post-processing for the current solution if needed.

After the solution has completed, the solver should send Cerberus an integer value indicating that the solution has been completed and to give some information on the potential convergence of the solution (as judged by the solver):

Direction	Name	Size and type	Content
S->C	NV	1*integer	A single integer 1, 2 or 3 (see below)
			1 = solution complete
			2 = solution complete and convergence reached
			3 = solution complete but diverged or other problems

correctPreviousStep

The solver may consider the current coupled solution converged and use this information to correct the previous time-integral that took the solver to the current time/time interval. If the solver cannot correct its previous time-integration or does not deem it necessary, it can simply ignore the signal.

Intended especially for burnup calculations but might be useful in some transient simulations also.

writeRestart

The solver should write its current state as a restart file. The path to the restart file may be fixed, given as a part of the solver input or as an input variable through Cerberus.

The contents of the restart file can be freely chosen by the developer and may be context dependent, e.g. on the simulation type that is being executed (stationary, burnup, transient etc.)

readRestart

The solver should set its internal state based on a restart file. The path to the restart file may be fixed, given as a part of the solver input or as an input variable through Cerberus.

Along with the writeRestart method, this one should allow e.g. for initializing a transient simulation based on a stationary state or continuing a previous calculation in another manner. The exact functionality can be chosen by the developer.

Field unit array

The field unit array consists of 5 integers (e.g.  1 -3  0  0  0 for mass density) specifying the dimension for each of the 5 SI base units for the field

No.	Property	Unit
1	Mass	kilogram (kg)
2	Length	metre (m)
3	Time	second (s)
4	Temperature	Kelvin (K)
5	Quantity	mole (mol)

This means that some of the basic fields that we might transfer will have types of

Field name	SI unit	Basic unit	Unit array
Mass density	kg/m3	kg*m-3	1 -3  0  0  0
Temperature	K	K	0  0  0  1  0
Power (integral)	W	kg*m2*s-3	1  2 -3  0  0
Power density	W/m3	kg*m-1*s-3	1 -1 -3  0  0