...
Source 5.0 also includes a range of upgrades and enhancements, as described below.
Date Range changes
Could this be under minor?
We have improved Date Ranges and fixed up some issues with their implementation. This will change results for a number of users, so we have implemented a Scenario Option to turn on the new Date Ranges algorithm.
Existing projects won't be using the new method and will have to manually swap over. This gives users a chance to test the new implementation before it becomes the only option.
In a later Beta release, we will implement a utility to convert across while keeping results consistent as much as possible along with user interface improvements to help users understand Date Ranges better.
Results Streaming
Results can now be streamed to files on disk rather than stored in memory. This is to prevent memory issues with large runs. Results streaming is off by default, however, it can be turned on in Project Options:
River Operations
We have been working with the MDBA to improve Source in operations mode. A focus has been enhancements to the Operations Tabular Editor. The editor is now more stable, easier to use, allows you to seamlessly add and remove nodes and is independent of the recorder selection.
NB: Geoff D to add a new picture
A new order recorder
We have developed a new recorder to better analyse orders, it shows all the nodes and links in a network and what their orders and requirements are on a particular date. Available under: Miscellaneous \ Ordering \ Order Crawler. The Order Crawler is a fully interactive zoomable schematic. Note that for large networks, it does record a lot of data and is primarily for problem-solving and not intended to always be turned on.
Results Streaming
Results can now be streamed to files on disk rather than stored in memory. This is to prevent memory issues with large runs. Results streaming is off by default, however, it can be turned on in Project Options:
The format used (called Source Time Series Database .sdb) is now also available as a general timeseries format.
...
Source.DataUtility.exe - eWater Source Time Series Data Utility Usage: -?, --help[optional]... Show usage help -i, --input[optional]... Input file to read timeseries from. All standard TIME timeseries formats are supported. Example: -i "Run1.sdb" or -i "Run1.res.csv" -o, --output[optional]... Output file to save timeseries to. The extension selects the output format based on the standard TIME I/O formats. A directory of the name specified will be created if the format only supports a single timeseries. Example: -o "output.res.csv" for Source Results CSV or -o "output.sdb" for SourceDB etc. -k, --key[optional]... Key of a timeseries to output. This can appear multiple times. If no keys are specified then all timeseries will be exported. Use the '--list' option to list timeseries in the input file. The DisplayName is used as the key.Example: -k "someResult" -k "someOtherResult" -q, --query[optional]... SourceDB filter query. Use the '--list' option to list timeseries in the input file together with possible MetadataKeys to filter on. Example: -q "WaterFeatureType=Storage" -q "ElementName=Storage Volume" --startTime, --st[optional]... Start time for extracting partial results. If not specified the start of the input timeseries is used. Example: --st 1/1/2011 --endTime, --et[optional]... End time for extracting partial results. If not specified the end of the input timeseries is used. Example: --et 31/12/2011 -l, --list[optional]... List the timeseries and metadata contained in the input timeseries file. Standard output may be redirected to a '.json' file. -a, --addOrReplace[optional]... Add input timeseries to an existing output SourceDb (.sdb) file. Existing timeseries with matching metadata will be replaced. Only supported for SourceDb output files. Example: --addOrReplace -f, --force[optional]... Don't prompt for confirmation when overwriting an existing file.
Performance improvements for the command line and Service interface
We have improved the performance of the command line so that less memory (RAM) is now used during a run. Run times have been reduced in the command line.
The command line now has additional options for results saving:
...
--resultsOutputMode
There are three modes currently available:
- NoOutput - Does not save results anywhere, hence does not fetch results either. Can be used where the result data files are ignored. E.g. in tests, or where a plugin gives a different pathway to results. This improves performance when using automatic exporting tools such as Scenario Options: Simulation results or Simulation Log which automatically exports results after each run to a specific location.
- ClientNamedFormat- Results fetched to client, saved to output file from there, in a format based on output file extension. The data is transferred back to the client and saved (current behaviour).
- ServerNamedFormat - Results saved to output file directly from server, in a format based on output file extension. Does not fetch results to client. This "server side" save is more efficient. Default for InProcess mode, since "server side" is local here. Can be used in Client mode, assuming the end user has access to the path provided to the server e.g. a network drive.
River Operations
The Operations tabular editor is now independent of the recording selection. You no longer get all recorders in the tabular editor. Recorders are now selected using the new column editor:
The tree on the left has also had major changes.
Resource Assessment Allocation table interface improvements
Could this be under minor?
You can now freeze columns and drag columns to reorder the columns in the Allocation table within resource assessment:
A new Water User and Constraint Phase
There has been a change to allow modellers finer control over function execution during the ordering process.
Phase logic has changed from:
...
- Evaluate functions and modelled variables - where Time of Evaluation equals Order Phase and where the function is not used at a network element.
- Process Water Users:
- Water User order calculations can be run in Parallel at this point in the ordering phase with the option enabled: "Process Water User Orders in Parallel": Scenario Options \ Performance
- For each Water User:
- Evaluate functions and modelled variables - where Time of Evaluation equals Order Phase, and where the function is used at this Water User.
- Calculate and distribute the demands of each Water user.
- Evaluate functions and modelled variables - where Time of Evaluation equals Order Phase, and where the function is used at this Water User.
- Constraint Phase:
- For each Network Element:
- Calculate minimum and maximum constraints at each element and pass the constraints downstream.
- For each Network Element:
- Ordering Phase:
- For each Network Element:
- Evaluate functions and modelled variables - where Time of Evaluation equals Order Phase, and where the function is used at this network element.
- Calculate regulated orders and pass upstream.
- Calculate off allocation orders and pass upstream.
- For each Network Element:
To:
...
10. Water User and Constraint Phase
- Evaluate functions and modelled variables - where Time of Evaluation equals Water User & Constraint Phase and where the function is not used at a network element.
- Process Water Users:
- Water User order calculations can be run in Parallel at this point in the ordering phase with the option enabled: "Process Water User Orders in Parallel": Scenario Options \ Performance
- For each Water User:
- Evaluate functions and modelled variables - where Time of Evaluation equals Water User and Constraints Phase, and where the function is used at this Water User.
- Calculate and distribute the demands of each Water user.
- Constraint Phase:
- For each Network Element:
- Evaluate functions and modelled variables - where Time of Evaluation equals Order Phase, and where the function is used at this network element.
- Calculate minimum and maximum constraints at each element and pass the constraints downstream.
- For each Network Element:
11. Ordering phase:
- Evaluate functions and modelled variables - where Time of Evaluation equals Order Phase and where the function is not used at a network element or is used at a Water User.
- Ordering Phase:
- For each Network Element:
- Evaluate functions and modelled variables - where Time of Evaluation equals Order Phase, and where the function is used at this network element.
- Calculate regulated orders and pass upstream.
- Calculate off allocation orders and pass upstream.
- For each Network Element:
The important thing to note is that the timing has changed.
Example: A function assigned to a Water User, that says it evaluates in the Ordering Phase, will now happen later than it previously did. To evaluate it at the same time as it did in v4.11, the time of evaluation needs to be moved to the Water User & Constraint Phase.
Migration Code
So to minimise the effect in splitting this phase in two, we've written some migration code.
The logic is:
- IF a function or modelled variable has a time of evaluation of 'Ordering Phase' (in v4.10.1)
- AND it is used on a model that is either:
- a Non-Network Element (e.g. Resource Assessment, Env Flow Manager)
- a Water User
- THEN move its time of evaluation to 'Water User & Constraint Phase'
also
- IF a function or modelled variable has a time of evaluation of 'Ordering Phase' (in v4.10.1)
- AND it is not used anywhere
- AND it has Force Evaluate set to On
- THEN move its time of evaluation to 'Water User & Constraint Phase'
otherwise, we leave it as 'Ordering Phase'
Limitations
We know our migration codes is not perfect:
- We don't cater for Functions activated via input sets.
- Functions not used on a model, but only used by other Functions will not be moved
Recommended steps
We recommend you review any functions that are evaluated in the ordering phase and ensure they are configured correctly using: Edit \ Feature Table.. \ Functions.
New Version Number system
We have changed the version number system to help users understand when they need a new version of Source to open projects. The previous database version was a letter that wasn't in the version number.
The previous system was:
...
Order
...
Name
...
New system:
...
Order
...
Name
...
Essentially a unique identifier for Source builds
For example, under the previous system the last beta release would have been 4.12.4beta, but under the new system it was 4.13.1beta
Summary of minor changes
...
We have introduced the concept of Locking to the Function Modelled Variables, selected variables can now be locked, making it less likely for mistakes to occur.
...
Ordering for the Irrigator demand model now includes an option to make the Flow phase and Order phase identical.
...
Data Sources can now find data based on the column name rather than the index. Being able to match by name is important when using Reload on Run. Match data items by Name is now the default option.
It will validate and fail to run if you are matching based on name and reload a file with duplicate names.
...
The Operations Tabular Editor now allows you to copy-paste into multiple cells and view charts directly in the Results Manager.
...
Constituent Trend and % removal model. Links and Storages decay models have been extended to include a decay to a background concentration rather than to 0 and a % removal that can occur after the decay.
...
Performance improvements
We have made changes to significantly improve performance for editing and saving in some large models, for one project this reduced the time to save from 12 minutes to 45 seconds.
We have also improved the performance of the command line so that less memory (RAM) is now used during a run. Run times have been reduced in the command line.
The command line now has additional options for results saving:
|
There are three modes currently available:
- NoOutput - Does not save results anywhere, hence does not fetch results either. Can be used where the result data files are ignored. E.g. in tests, or where a plugin gives a different pathway to results. This improves performance when using automatic exporting tools such as Scenario Options: Simulation results or Simulation Log which automatically exports results after each run to a specific location.
- ClientNamedFormat- Results fetched to client, saved to output file from there, in a format based on output file extension. The data is transferred back to the client and saved (current behaviour).
- ServerNamedFormat - Results saved to output file directly from server, in a format based on output file extension. Does not fetch results to client. This "server side" save is more efficient. Default for InProcess mode, since "server side" is local here. Can be used in Client mode, assuming the end user has access to the path provided to the server e.g. a network drive.
A new Water User and Constraint Phase
Improvements have been made to input more information around when constraints happen, to make information available earlier and finer control over function execution during the ordering process.
Phase logic has changed from:
10. Ordering phase:
|
To:
10. Water User and Constraint Phase
11. Ordering phase:
|
The important thing to note is that the timing has changed.
Example: A function assigned to a Water User, that says it evaluates in the Ordering Phase, will now happen later than it previously did. To evaluate it at the same time as it did in v4.11, the time of evaluation needs to be moved to the Water User & Constraint Phase.
Migration Code
So to minimise the effect in splitting this phase in two, we've written some migration code.
The logic is:
- IF a function or modelled variable has a time of evaluation of 'Ordering Phase' (in v4.10.1)
- AND it is used on a model that is either:
- a Non-Network Element (e.g. Resource Assessment, Env Flow Manager)
- a Water User
- THEN move its time of evaluation to 'Water User & Constraint Phase'
also
- IF a function or modelled variable has a time of evaluation of 'Ordering Phase' (in v4.10.1)
- AND it is not used anywhere
- AND it has Force Evaluate set to On
- THEN move its time of evaluation to 'Water User & Constraint Phase'
otherwise, we leave it as 'Ordering Phase'
Limitations
We know our migration codes is not perfect:
- We don't cater for Functions activated via input sets.
- Functions not used on a model, but only used by other Functions will not be moved
Recommended steps
We recommend you review any functions that are evaluated in the ordering phase and ensure they are configured correctly using: Edit \ Feature Table.. \ Functions.
New Version Number system
We have changed the version number system to help users understand when they need a new version of Source to open projects. The previous database version was a letter that wasn't in the version number.
The previous system was:
Order | Name | |
|---|---|---|
| 1 | Major Version number | |
| 2 | Prod/Beta identifier | All Prod numbers are even. All Beta identifiers are +1 the most recent Prod identifier |
| 3 | Regular Functionality Release | Incremented when a new Beta is released every 1 to 2 months. Rarely updated for Prod releases. |
| Not included in version number | Database Upgrade letter | Needed when changes are made that require DB upgrades between official releases. This is used to identify what the current database structure of the project will be. Resets to 'a' when the 'Regular Functionality Release' changes. |
| 4 | Build number | Essentially a unique identifier for Source builds |
New system:
Order | Name | |
|---|---|---|
| 1 | Major Version number | |
| 2 | Prod/Beta identifier/Regular Functionality Release | Reset when the Major Version number increments. Increments with every official Prod and Beta release. We may increase by more than +1 for production releases. |
| 3 | Database Upgrade Number | Needed when changes are made that require DB upgrades between official releases. This is used to identify what the current structure of the database's project will be. |
| 4 | Build number | Essentially a unique identifier for Source builds |
For example, under the previous system the last beta release would have been 4.12.4beta, but under the new system it was 4.13.1beta
Summary of minor changes
- We have made improvements to Date Ranges. This will change results for a number of users, so we have implemented a Scenario Option to turn on the new Date Ranges algorithm. Existing projects won't be using the new method and will have to manually swap over. This gives users a chance to test the new implementation before it becomes the only option.
We have introduced the concept of Locking to the Function Modelled Variables, selected variables can now be locked, making it less likely for mistakes to occur.
Ordering for the Irrigator demand model now includes an option to make the Flow phase and Order phase identical.
Data Sources can now find data based on the column name rather than the index. Being able to match by name is important when using Reload on Run. Match data items by Name is now the default option.
It will validate and fail to run if you are matching based on name and reload a file with duplicate names.You can now freeze columns and drag columns to reorder the columns in the Allocation table within Resource Assessment.
Constituent Trend and % removal model. Links and Storages decay models have been extended to include a decay to a background concentration rather than to 0 and a % removal that can occur after the decay.
- Annual Accounting recorder changes. Some annual accounting recorders at RAS system level were giving incorrect values. The recorded values of "Total account spill reduction", "Total usage water year" and "Total Account Adjustments" at RAS system level values will change. See RM-18149 for details.
- Ownership adjustment for storages and weirs - End of day volumes in Storages and Weirs can now be adjusted.
- Custom charts now include a search bar and more menu items under right-click. Customs charts are also no longer saved by default. All automatically generated charts are created under Temporary and only saved when moved into Saved. Users can also create folders and subfolders for organising custom charts.Urban Developer Recorders have all been converted to the new more stable recording framework. Allowing the inclusion of long descriptions for the recorders in Urban Developer and the Urban Developer Demand Model that is used in Source.
Community Plugin Changes
We have moved to .net 4.8.0 which requires Visual Studio 2019. We are planning on moving to .NET 5. The transition to .NET 5.0 won't be a quick process as some components we currently use will no longer be supported, however it does provide a potential pathway to running natively on Linux. Source can currently be run on Linux under Wine, but only natively on Windows.
...