Prophet Enterprise (PE)

• Performance enhancements
  • Improvements to the time taken to import a PBUZ file
  • Improvements to the time taken to display of lists of large workspaces n Improvements in the interaction between job templates and jobs
• User interface
  • Increased control over run structures and run settings
  • Improvements have been made to the backup, restore and promote operations
  • Improvements have been made to role permissions

• Flexible Results: Aggregation results
• Flexible Results: Individual policy results
• Flexible Results: Flexible Results API
• Table and model point optimisations with modules
• Usability enhancements

• Flexible Results: Distributed Database Output
• Results change
• Looped Modules
• Advanced Vector Extensions (AVX)
• Prophet Grid Administrator API
• Improved data transfer protocols
• Size of the batch can be specified for split by model point batch
• Categories of finished jobs
• Improved security
• Streamline machine entry
• Improved error logging

• Improvements to speed up workspace import from a PBUZ file, for workspaces with many products and accumulations. On average users should notice 40% shorter times to import, and even 90% shorter for workspaces with a massive number of products and accumulations.
• A redesign of the way the user interface displays products and accumulations together with the workspace. This leads to a much quicker response time to display of lists of large workspaces.
• A user interface performance improvement for Prophet Enterprise systems with large numbers of parameter definitions. As a result, a user may observe quicker times to display lists of finished jobs and active jobs. We have observed improvements of up to 90% for finished jobs and 50% for active jobs.
• Improved interactions with job templates and jobs. Now times to open, edit or save these entities are at least 75% shorter. The time needed for job submission has also decreased by 15% on average and even more for submission of multiple jobs at the same time.

All of the improvements impact the responsiveness of Prophet Enterprise API calls like: Create Job Template, Edit Job Template, Create Job, Get Job, Edit Job, Submit Job but also Get All Workspaces, Get Active Jobs and Get Finished Jobs.

User interface enhancements
In Release 2021 Q2 Service Pack 9 we have made improvements across several parts of the user interface:

Run structures and run settings
With Release 2021 Q2 Service Pack 9 a Prophet Enterprise user with appropriate permissions will now be able to create a new run structure or run setting and copy these two entities between different workspaces. New options are available in the user interface and via the Prophet Enterprise API (using either C# or PowerShell).

These changes could save time and effort when importing new workspaces; promoting workspaces as well as performing Prophet Enterprise workspace updates. This improvement also enables users to write more powerful Prophet Enterprise API code for actuarial model changes.

Backup and restore
We’ve improved the robustness and stability of backup, restore and promote operations. The enhancements include: n Introduction of a temporary backup file to prevent the file being copied too early.

• A series of improvements for backing up and restoring .amb files with large entities.
• Displaying a progress bar when the system validates dependencies for selected entities.

Role permissions
Unlocking entities in Prophet Enterprise Studio has historically been restricted to users with the Super Administrator role only. For this release, there is a new permission which allows the ability to unlock entities to be granted to other users who are not Super Administrators.

Prophet Enterprise makes use of entity locks to ensure that conflicts are not created when multiple users try to change the same thing. The locks are usually automatically released when the operation completes. It is possible, however, for the lock to not be released properly and in this circumstance, historically, only a Super Administrator or the user themself would have been able to unlock the object.

The new permission is a process access permission type, which can be granted to any user with a default or custom role. The unlock operation can be used on any locked entities and works irrespective of who originally locked the entity. Therefore, we recommend granting this permission to any admin user who may need to control user access and entity creation in the environment.

Prophet Enterprise API

• With Release 2021 Q2 Service Pack 9 two new commands: DeleteVersion and DeleteChildVersion have been added to the Prophet Enterprise API. These can be useful for cleaning up a Prophet Enterprise environment where many versions and sub-versions have been created over time and are no longer required.
• Prophet Enterprise API users can now perform a backup, restore, and promote with the Execute on server option enabled. The new parameter is optional and is turned off by default.
• Prophet Enterprise API users are now able to upload all files under a workspace in the same way as the existing user interface functionality.
• There is a new method which allows Prophet Enterprise API users to delete a workspace version.
• The Prophet Enterprise API documentation has been updated to show all available methods.

PEGAdmin API

• Users with the PEGAdmin API licence can now check the status of the Prophet Enterprise Monitor Manager service via the API. API commands have also been included to start, stop and restart Monitor Manager.

Calculation Engine

• We now support Prophet code compilation using the 2019 edition of the Microsoft Visual 2019 C++ compiler. We no longer support compilation using the 2015 edition.

Plugins

• It is now possible for Prophet Enterprise parameters to be used in plugin related (Intex and Andrew Davison specifically) run settings. This new enhancement allows a user with Submit job permissions – but not with either Edit job template or Edit job permissions – to set the value on job submission. This enables users to change the plugin value or locations at submission without altering the job template.

Documentation enhancements

• The Prophet Enterprise Technical Guide has undergone a thorough review to bring it into line with latest guidelines for Prophet Enterprise installation. The guide is now more comprehensive with the steps required for setting up a fully functional environment laid out chronologically rather than grouped based on Prophet Enterprise component.

Licensing
The following improvements have been made to the Prophet licensing technology:

• In the Client Licence Server (CLS) there is new Export functionality, allowing you to export the filtered on-screen data to a csv file. You can export data from the following tabs: Licences, Licence Consumptions, Synchronisations, Users and the Clients details tab (for a specific client machine).
• CLS users now have roles. User permissions have been split into two roles – Administrator and Licence Administrator – allowing you to better manage CLS user access and entitlements.
• The Administrator role is for those users responsible for server maintenance, ensuring its connectivity with both the FIS Licence Server and Prophet users. These users also have permission to create and delete CLS users.
• The Licence Administrator role is intended for those users tasked with administering the licences themselves, for example, adding users to a group licence and looking at licence consumption history.
• In the CLS portal, users now have the option to remove multiple group licence users in one go from the group licence Users tab. Previously, users would have to be removed separately.
• A new option has been added to the CLS installation giving you the ability to change the CLS hostname to the fully qualified domain name. Previously the hostname was set to the ‘simple’ name of the server on which the CLS was being installed by default.
• You can now view a detached licence’s products and features on the detached tab in Prophet Licensing.
• The CLS user documentation is now provided as HTML5 help embedded in the CLS portal, to make the documentation much easier to access. The help is completely accessible from within the web browser, with no read access to the CLS installation directory required.

• Projection text results (PRN files) were written twice; by the workers after each product and again by the taskmaster at the end of the run pair/job. This could lead to longer run execution times. They are now only generated by the workers.

Prophet Enterprise 2021 Q2 Service Pack 7

• An issue of being unable to view any variables in Module (Results) Diagram View for pre-decision module links has been resolved.
• There was an issue with PMPP mode running significantly slower than non-PMPP when the model was making intensive use of READ_RESULTS function due to a critical section in the error handling logic. This has now been optimised and will also benefit models that make intensive use of table reading functions and plugin calls where similar logic applied.
• The default logging settings for errors and warnings were found to be not persisted from Prophet Professional to Prophet Enterprise. This has been fixed.
• An issue of base set status not correctly showing in the base set document panel when the user updated documents and templates linked to a base set has been fixed.
• When the user enters Edit mode on an Assumptions Manager job definition, the sensitivity document that is selected will now remain selected.
• An issue of reporting date parameter selection throwing an exception on basis document submission has now been resolved.
• Workspace list is now sorted in the dropdown box of the Assumptions Manager Job Submission dialog box.
• A reduction in frequency of validating AD user per web request has been implemented.
• There was an issue where if you tried to open a very large finished jobs list, Prophet Studio could become unstable or time out. This issue has been resolved, and accessing large finished jobs lists in PE API has also been improved.

Prophet Enterprise 2021 Q2 Service Pack 5

• System logs were not created. We have fixed that so they now can be seen in PEGAdmin tool.
• Users whose client and server machines used different time zones would see a delay in the yellow banner being displayed in the basis document after updating an underlying base set. This is now fixed.
• The Assumptions Manager report, which is produced with an Assumptions Manager job or when a user generates PP data, was showing incorrect reporting dates for unlinked tables being used in the job. The issue has been fixed.
• We have added information on the version of tables to the Assumptions Manager report. It is now consistent with the information in the basis document version used for the Assumptions Manager job. The change affects the Assumptions Manager report produced with an Assumptions Manager job and when the user generates PP data.

Prophet Enterprise 2021 Q2 Service Pack 3

• There was an issue in which read results in a run failed. This was caused when devices or resources were busy, and resulted in multiple errors in the run log as well as the run failing. This issue has been addressed, and the run completes successfully.
• There was an issue whereby Excel Reporting could not access results when using copy results functionality and when the model contained modules. The error reported an issue with a missing metadata file. This has now been fixed allowing such files to be read correctly using Excel Reporting. Note this issue has also been fixed in 2020 Q2 Service Pack 4.
• There was a problem where Prophet Enterprise would report and use an incorrect number of cores on a worker machine. This has been fixed.

Prophet Enterprise 2021 Q2 Service Pack 2

• There was an issue installing the Client License Server because some machines didn’t allow the user to elevate PowerShell execution privileges. This has been fixed in the CLS installer.
• It is now much quicker to open and edit basis documents. The improvement is also noticeable for basis documents updating with yellow banner to apply any base sets’ changes.
• Run logs for results stored to Arango Database will now open correctly. The change was made to the Open Run Log option to handle both the job path and the results path typed into the Select results directory dialog box.
• There was an issue for workspace update in cases where the user deleted some or all of the accumulations in Prophet Professional prior to the update to Prophet Enterprise. In such case related run structures were corrupted. This has been fixed.
• There was an issue with updating basis documents with the Assumptions Manager API if there were links to Input Data – Excel Spreadsheets. This has been fixed and the changes are applied with the Set-BasisDocument command.
• The performance of workspace import has been improved. The benefit should be observed mostly for environments with large base sets data.
• The performance of submitting Assumptions Manager job definitions has been significantly improved. The change was done for better handling submission of multiple job definitions at a time.
• Assumptions Manager job definitions will now open much quicker in read only mode. With this change only necessary information is being called on job definitions opening.
• There was an issue with persistency of the AoC check box in Assumptions Manager job definitions on backup and restore operations. This is now fixed and the check box selection is correctly restored or promoted.
• An issue where metadata in the FRAPIResultsSource.GetRunNumberList() method was returning default values for jobs with a failed status has been fixed so that data is returned if available.
• An issue has been fixed for nested structure jobs with an inner structure split by simulation run setting where the machines needed for the inner structure were being released prematurely as they weren’t required for the outer structure, resulting in the inner structure using fewer machines than requested.

Prophet Enterprise 2021 Q2 Service Pack 1

• The Prophet Enterprise Technical Guide has been updated to describe how to perform an upgrade of the Prophet Enterprise environment. This section was not present in the documentation for Release 2021 Q2.

Prophet Enterprise 2021 Q2
Performance improvements

• The base set entity data code has been improved so that the latest version of base set tables is returned. It was previously returning all versions.
• Run settings will open quicker as the validation of base set links has been improved.
• The basis document update has been improved by only checking the uniqueness of the entity name against the original name stored in the basis document data, instead of all entities from the database. n Finished job and job templates now open faster.
• The performance of duplicating a run structure has been improved.

Backup/Restore and Promote

• An issue whereby users could not restore a large base set from a backup file has been fixed. This issue was found when restoring a base set with a few hundred tables where each table had lots of versions.
• After promoting a workspace from one environment to another there was an issue with comments in the audit log being removed or blanked out. This has now been fixed.
• There was an issue when if a user cancelled a backup, the backup would incorrectly continue. This has been addressed and the backup is now stopped. A similar issue where an object restoration with the Execute on Server option enabled could not be cancelled has also been addressed. The cancel operation is now properly logged in the domain log.
• Previously, when taking a backup of any Prophet Enterprise object, all the categories which could be applied to that object would be automatically included in the backup. This was the case for all categories regardless of whether their values were assigned to the object or not. This has been fixed so that now only categories which are assigned to an object are included in the backup.
• Promoting more than one dataset at once resulted in only one completing successfully. This has been fixed.

User interface

• Trailing slashes in the Web Service URL when logging into Prophet Enterprise previously caused an Error in web service call which has now been fixed.
• If a user specified more than one parameter in the job template then only the first parameter would be shown in the Parameters tab when the user selected the Show only parameters that are in use check box. This has now been fixed.
• If a parameter took a value that has a file extension (such as a table name), when the Show only parameters that are in use check box was selected the parameter would not appear in the list. This has been fixed.
• An issue whereby changes to the Text Results selection in the run setting was not saved if the user exited Edit mode has been fixed.
• Previously a user was unable to view stochastic summary results from within Prophet Enterprise Studio because the Stochastic Summary result type was not available on a dropdown list. This has been fixed.
• There was an issue with the Disable Multi-Threading information shown when your job used parallel model point processing (PMPP). When a user selected a job from the Active Jobs list, the Properties pane was displaying Disable Multi-Threading as True instead of False. This has been fixed.
• In previous versions a user would get an error message when clicking View Results on a cancelled job, even though there are no results for a cancelled job. The issue has been fixed so the error no longer appears on clicking View Results but you are unable to run any query.
• If a user created a version or reverted to a version on a workspace with structure links, those structure links would be missing. This has been fixed.
• If a user used the Export Results functionality to export results and subsequently reimported those results using the Import Results functionality, opening the runlog would fail. This issue has been fixed.
• An issue where, after upgrading Prophet Enterprise from 2018 Q3 system version or earlier, existing runlogs could not be opened from Prophet Studio has been addressed.
• A workspace would not upload into Prophet Enterprise if it contained structure links that were not used by any of the run structures. This has been fixed.
• Users could not see data sets in the Prophet Enterprise Explorer, even if this view was selected for their role. This issue has now been fixed.
• There was an issue with reimporting a workspace such that if a user removed any accumulations in Prophet Professional but they were still selected in some run structures in Prophet Enterprise, the workspace import failed if the structures were not included in the workspace import. This has now been addressed.
• An issue with changing compartments for a workspace, which references any base set, has now been fixed.
• An issue opening aggregation results in External Diagram View such that the results type was left blank has now been resolved.

Prophet Enterprise API

• Model point batch size should only be within the range of 1 – 1,000,000. If a user tried to set the value outside of this range in the Prophet Enterprise API then the user would not be informed of that and the API would just assign the range boundary value automatically. In this release, if a user assigns a model point batch size outside the permitted range, a warning message is triggered and the previous value is kept.
• An issue where a user was not able to submit a job with integer numeric parameter values has now been fixed.
• An issue where a user was not able to overwrite parameter values in a job template has been addressed.
• A small enhancement has been made to the Prophet Enterprise API so that a user will be able to upload all files under a workspace via the API.

Push to PE

• Changing a machine group for a queued Push to PE job would cause the job to fail. This has been fixed.
• There was an issue with the Push to PE feature such that when submitting multiple run pairs, the second and subsequent job run pair would fail. This issue has been resolved, with the following limitations in Push to PE run pairs. For a specific workspace:
  • Run structures must be unique.
  • Run numbers must be unique.

The preparation stage for each of the run pairs is performed sequentially rather than in parallel. The runtime of multiple run pairs can be performed in parallel. You cannot suspend and resume a Push to PE job. Instead, cancel the job and resubmit it once the other jobs for that workspace have finished.

Calculation engine

• A couple of issues obtaining the results metadata of products in certain types of Prophet Enterprise run, such as when using Copy Results functionality where the destination already has results in it, has been fixed. The issue caused the results to be unreadable for affected products and affected both READ_RESULTS during a run and Excel Reporting.
• When submitting a job with a READ_RESULTS call reading from an external results location, where those external results were produced using copy results the read would fail in the case where the copy results location contained results for multiple jobs with different variables in the results data. This has been addressed.
• In the 2019 Q2 and 2020 Q2 releases, if you used a parameter to name your run number folders in combination with the copy results functionality then pre-existing folders in the specified location would be overwritten, even if different parameter values for the run number folders were used. This has been fixed.
• An issue where PMPP runs could fail due to a clash between the calculation threads trying to read results metadata and the thread outputting finalised spcode results has been addressed. The issue was first noticed in a job using READ_FLEXIBLE_TABLE.
• A bug where runs using the Calculated variables run structure option, in conjunction with Table and MP File Optimisations (TMPO) set to Full, could erroneously produce missing DLL errors has been fixed. This was due to the Same As products trying to load a DLL with the name of the base product despite TMPO=Full runs producing product DLLs for each Same As product and thus not producing a base product DLL.
• An issue in which selecting Only run first model point for each Sub-Product Code on the run setting did not always persist their values between dynamic loop iterations in a PMPP run has been addressed.
• A bug where accumulations could not be generated for a Prophet Enterprise job, where the accumulation read from model components in multiple dependent jobs, has been fixed.
• There was an issue when writing results for split by simulation runs to ArangoDB, where each simulation could write out the same unit prices file, which caused a conflict on writing results. This issue has now been resolved.
• For a job with read functions – PROJ_RESULT, STOCH_RESULT or AGG_PROJ_RESULT – the read would unexpectedly fail if the path in the external results location provided in the run setting included the run number folder. In the case where the run number folder matched the run number specified as a parameter value in the read function, the read would fail. This issue has now been resolved.
• In the 2019 Q2 and 2020 Q2 releases, if a job was reading from an external results location (provided in the run setting) using either a READ_RESULTS function or an accumulation, the run number from which the value was read from would differ depending on the system version used to produce the original results. Both READ_ RESULTS and accumulations have been corrected to always read from the run number corresponding to the run number being processed in the current job, regardless of the results version.
• In the 2019 Q2 and 2020 Q2 releases, the time taken to create the zip file when the create a results file for download option was enabled could be noticeably longer than in earlier releases. This has been resolved by changing the compression library used and users can now expect the post-job zipping time to be as good as, if not better than, previous levels.
• A bug in which reading outside array bounds failed in descending optimised loops for 2-dimensional arrays has been fixed so that zeroes are returned outside the array bounds.
• A bug that could cause structured exceptions when an extended formula array resized while being read has been fixed.
• A bug has been fixed where some looped module inputs, that vary by loop iteration and are used directly in IF conditions, were not updated for each loop iteration within their dependent module formulae.
• The error handling of the ARRAY_MAX function has been improved to prevent structured exceptions in some cases, and make clearer where the syntax is incorrect in other cases.
• A syntax-checking bug has been fixed that caused incorrect formulae, with respect to data types and comparisons, to not be caught properly. This resulted in malformed C++ code generation and a “merged code error” with C++ compiler messages that would not be understandable in terms of the underlying fault in the model’s definition.
• A t-limiting bug has been fixed that caused errors involving uncalculated values and out of bound time references to be masked and zeroes silently used instead.
• A t-limiting issue affecting t-independent extended formula, sequenced in the same calculation group as variables that are not t-limited, has been fixed. This issue had caused extended formulae variables to appear in the DGN but not actually be executed, and was observed due to incorrectly ignored breakpoints in the Prophet Debugger. Perturbing the sequencing by adding in references to the affected extended formula had been the only workaround prior to the fix.
• An update to the 2017 and 2019 compilers, to fix a C++ defect, caused existing Prophet code generation to start failing. Prophet code generation and compiler flags have been altered to comply with the new rules.
• The validation check that ensured that each data row in a model point file had the same number of columns as the header, after allowing for treating consecutive column delimiters as one column delimiter, was only working on the first data row. The check has now been fixed for all the model point data rows.
• An inefficiency that caused empty or zero filled results files to be produced for unexecuted products, when a run is aborted, has been fixed.
• An issue where lots of small files and empty subfolders were left behind in the Prophet temporary directory after a run finished has been fixed.
• We have addressed an inefficiency in the handling of dynamic debug results files in order to reduce the number of unnecessarily repetitive file operations. In addition, the code for creating a sub-directory, whose parent folder(s) may not exist yet, has been optimised to further reduce file operations.
• The logic for moving or copying a file over the top of an existing file was inefficient and could fail if the file server did not delete the existing file in a timely fashion, for example due to anti-virus scanning. This issue has been addressed.
• A performance inefficiency in READ_RESULTS has been fixed that particularly affected products that read a lot of results such as Summary products or ALS External.
• An inefficiency in the results formats introduced in Release 2019 Q2, that caused the results to include unnecessary data repetitively, has been addressed. The issue was most noticeable when using stochastic results on the file system where the number of variables and time periods per simulation was low. However, the issue still represented an inefficiency for other results types such as projection, especially when variables and time periods were heavily trimmed by the user.
• An issue causing runs to display “invalid date/time string” warnings in the run monitor has been addressed. The issue was compiler-related and only affected certain regional settings in a non-deterministic manner.
• In the previous release, if you changed the selected variable group on an aggregation results definition post-run then every product selected in the definition was regenerated. This has been fixed such that only the products affected by that change in variable group are regenerated.
• An issue where certain results types, when using the file system for results storage, are appended to for each dynamic iteration rather than replaced, has been addressed. This caused certain results to grow significantly bigger than pre-9.1 results. To fix this issue, t-independent variables are only output in the first dynamic loop for aproj, astoch, pproj and pstoch. This also prevents additional issues where t-independent cumulative variables, including ones that do not vary by dynamic loop start date, effectively reduced significantly as policies expired across dynamic loop iterations, and where the sum across aggregation values might not agree with the totals when one or more aggregation values ended up with no more active policies left to process prior to the end of the dynamic loop. These latter issues still exist for legacy results types, with respect to spcodes, such as projection and stochastic.
• An inefficiency with the READ_RESULTS function was introduced in Release 2019 Q2 where the Calculation Engine (CE) checked the results frequently to identify whether a requested result was part of a product or accumulation. This involved checking whether the requested product or accumulation was part of the currently executing run pair, and if not, checking whether it was part of the pre-existing workspace results, which is relatively expensive in terms of run time. Performing this check repeatedly for every READ_RESULTS call could significantly impact the runtime of products using lots of READ_RESULTS calls, for example Summary library products. If the workspace was located on a remote location to the worker(s) then the performance impact could be more adverse. This issue affected models where a lot of products and accumulations results being accessed were either missing or not part of in the external results location. This issue has now been fixed.
• Prophet 2019 Q2 introduced an issue in which all variables for a particular set of variable-by-time results were loaded into the results cache when a single variable was requested. The issue only affected the CE results-reading runtime functions PROJ_RESULT and STOCH_RESULT and the Excel Reporting worksheet functions of the same name. This could cause a significant memory increase during runs which access a small subset of variables across a large number of these sets, for example a summary product accessing a small set of variables across a large number of products and sub product codes using the affected functions. This issue has now been addressed for results stored using the file system directly, but not for results stored using ArangoDB.
• There was an issue with sensitivity documents if the sensitivity value for an assumption was equal to the value defined in the related basis document, and the value in the basis document was defined to be different from the underlying table value. Under such conditions the sensitivity resulted in the underlying table value being used instead of the value defined in the basis document. The issue was present for both cases, when applying sensitivity to the reporting date and to a specific AoC step. This has now been fixed.
• Two threading issues in the Prophet 2020 Q2 Calculation Engine that affect PMPP have now been fixed:
  1. If using PMPP and READ_FLEXIBLE_TABLE it was possible to get a putResults error of the type “concurrent or
     recursive” calls.
  2. If using PMPP and modules it was possible to get a structured exception when two DLLs both tried to load the same table at the same time. Depending on whether fast runtime or fast compile was used in the structure, or whether a module link was set to be looped, a DLL could represent a single module shared amongst module links in the same product or individual module links.

Grid services

• Scheduler and Monitor Manager services in Prophet Enterprise will now show messages in their log files to record when they are started. This can help identify when they have been restarted either deliberately or by the operating system due to a fatal error.
• The configuration file – system.config.xml – was not accessible from Prophet Grid Administrator (PEG Admin). This issue has been fixed so that the configuration file is now editable.

Licensing

• Validation has been inserted to prevent detached licence users from changing their Client Licence Server (CLS) configuration details in Prophet Licensing, as this renders the licence unusable.
• At CLS installation, in cases where startup of the CLS Windows service took longer than 90 seconds, the installation could fail with error code 1920. A change to the CLS installer should mean this error no longer appears.
• Robustness of the CLS Windows service has been improved by removing a database shrink operation.
• Email verification in the CLS has been improved such that you can no longer add duplicate user email addresses to the same group licence.
• There was an issue recording licence consumptions in the CLS portal when the CLS machine’s disk was full. This has been fixed so that now if licence acquisition works then the consumption is correctly rendered in the CLS portal.
• Some settings changes in the CLS require a restart of the CLS Windows service to take effect. However the message informing CLS users of this was easy to miss. The instruction to restart is now clearer.

Plugins

• An inefficiency in the Intex plugin has been fixed so that we no longer request the underlying Intex API produced output that can’t be accessed by a call to the Intex plugin.
• A missing ADCO DLL that was only needed when using ADCO implicitly with Intex, via the INTEX.DEAL function call, has been added back into ADCO Plugin version 2.2.0.
• Intex has dropped support for calling ADCO directly from within their software to get prepayment rates. This functionality no longer works and resulted in a structured exception in the Intex plugin for versions 3.4g onwards. The Intex plugin now does a check to error gracefully if you try to use the ADCO prepayment method when calling INTEX.DEAL. Version 3.4f_p2_sp1 is the last Intex plugin to support this method. We are working on an alternative approach for a future release.
• A bug has been fixed in the Intex plugin affecting the PAY_TO_PAY function when requesting certain results types such as WAC, WALA or collateral balance ratios. The problem was that a weighted average calculation did not divide by the sum of the weights at the end.

Excel Reporting

• An issue involving the time taken for Excel Reporting to open projection results has been fixed.
• In previous versions, on the Prophet tab of the ribbon in Excel, if you clicked Options and then cleared the default selection of Show Variable Errors, an “unset nullable” error was returned instead of zero. This issue has now been resolved.
• An issue has been fixed involving Excel Reporting being unable to get metadata for Same As products in Prophet Enterprise copy results folders when folders contain existing results.
• An issue has been fixed involving Excel Reporting queries not executing properly if worksheet functions were called first and the Results subfolder was passed in as the result store location to the query.
• There was an issue involving the time taken for Excel Reporting to respond when unable to connect to ArangoDB. This issue has been fixed.
• An error in Excel used to appear when using the Results_Date_Time function to obtain unit prices results in ArangoDB. This is no longer an issue.

Aggregation results
Aggregation results allow you to generate projection results aggregated by a model point variable of your choosing, called the aggregation variable. Aggregation projection results can be created for up to two aggregation variables per product in addition to existing sub-product aggregation functionality.

Aggregation results are specified in a new workspace object, the Aggregation Results Definition, before adding to the run setting in the Flexible Results section of the Results tab.

The new results definition can also be created, edited and added to your job’s run settings via the API.

Aggregation results output can be viewed directly in the Prophet Studio application, Results Viewer and with Excel Reporting through a new AGG_PROJ_RESULT function.

Individual policy results
Individual policy results were previously only available to clients with a Production Interface licence, they are now also available to users licensed for Flexible Results. Individual policy results allow you to create projection or stochastic output for specified policies in your model point file, up to a 100 policy limit.

The specification of these results adopts a similar approach to aggregation results, via a workspace object which can then be referenced by multiple run settings. Individual policy results can also be set up via the API, read directly in the Prophet Studio application, Results Viewer and with Excel Reporting through a new POLICY_RESULT function.

For further information on either aggregation or individual policy results, see the Prophet Professional 2020 Q2 Bulletin.

Flexible Results API
The Flexible Results API (FR API), which was released in 2020 Q1 for Prophet Enterprise users, has been updated to read both aggregation and individual policy results. The FR API is a .NET Dynamic Linked Library (DLL) that can be used by a bespoke application to directly read Prophet results at scale without the need to first load the data into the Prophet Results Database (PRD).

The Insurance Data Repository (IDR) can now use FR API to efficiently extract, transform and load Prophet results data with minimal changes to any existing jobs that were using PRD as a data source.

Table and model point optimisations with modules
Table and model point optimisations can now be used with products containing modules. Table and model point optimisations can be enabled on the Runtime tab of the job or job template.

For further enhancements made to modules as part of this release, see the Prophet Professional 2020 Q2 Bulletin.

Usability enhancements

• Improvements have been made to speed up the access and display of results in the Prophet Enterprise user interface.
• The performance of opening run settings when they reference base sets in the tables tab has been improved.
• Domain settings have been split between two permissions: Administrative Tasks and Actuarial Administration. The new Actuarial Administration permission grants access to non-system administrator permissions like the ability to amend table locations, job priorities or default upload options. The change does not affect existing roles but does enable custom roles to be more specifically tailored to your environment’s needs.
• Monitor Manager can now update itself when the list of machines in a grid changes without needing to be restarted. It does this by polling the list of machines, and when it detects any changes it adds and removes machines as necessary. The polling rate is configurable but by default is set to 0 so that no polling happens and the original Monitor Manager behaviour is preserved.

• There was an issue with the Push to PE feature so that when submitting multiple run pairs, the second and subsequent job run pair would fail.

  This issue has been resolved, with the following limitations in Push to PE run pairs. For a specific workspace:

  • Run structures must be unique.
  • Run numbers must be unique.

  The preparation stage for each of the run pairs is performed sequentially rather than in parallel. The runtime of multiple run pairs can be performed in parallel.

  You cannot suspend and resume a Push to PE job. Instead, cancel the job and resubmit it once the other jobs for that workspace have finished.

• The Prophet Enterprise functionality that lets a user copy results to an output location, and potentially merge them by run number with existing results in that location, did not correctly work if the location had existing results in it. In such situations, results viewing components such as Excel Reporting could not report on SameAs products in the results: an error, relating to not finding a key piece of metadata required to interpret the results files, was returned instead of the requested results. This issue has now been fixed.
• An inefficiency with the READ_RESULTS function was introduced in Release 2019 Q2 where the Calculation Engine (CE) checked the results frequently to identify whether a requested result was part of a product or accumulation. This involved checking whether the requested product or accumulation was part of the currently executing job, and if not, checking whether it was part of the configured external results location(s), which is relatively expensive in terms of run time. Performing this check repeatedly for every READ_RESULTS call could significantly impact the runtime of products using lots of READ_RESULTS calls, for example Summary library products. If the results were located on a remote location to the worker(s) then the performance impact could be more adverse. This issue affected models where there was a valid external results location configured and a lot of products and accumulations results being accessed were either missing or in the external results location. This issue has now been fixed.
• In previous versions, on the Prophet tab of the ribbon in Excel, if you clicked Options and then cleared the default selection of Show Variable Errors, an “unset nullable” error was returned instead of zero. This issue has now been resolved.
• Prophet 2019 Q2 introduced an issue in which all variables for a particular set of variable-by-time results are loaded into the results cache when a single variable is requested. The issue only affects the CE results-reading runtime functions PROJ_RESULT and STOCH_RESULT and the Excel Reporting worksheet functions of the same name. This can cause a significant memory increase during runs which access a small subset of variables across a large number of these sets, for example a summary product accessing a small set of variables across a large number of products and sub product codes using the affected functions. This issue has now been addressed for results stored using the file system directly but not for results stored using ArangoDB.
• There was an issue with sensitivity documents if the sensitivity value for an assumption was equal to the value defined in the related basis document, and the value in the basis document was defined to be different from the underlying table value. Under such conditions the sensitivity resulted in the underlying table value being used instead of the value defined in the basis document. The issue was present for both cases, when applying sensitivity to the reporting date and to a specific AoC step. This now has been fixed.

Prophet Enterprise 2020 Q2 Service Pack 1

• This update fixes two threading issues in the Prophet 2020 Q2 Calculation Engine, that affect PMPP:
  1. If using PMPP and READ_FLEXIBLE_TABLE it is possible to get a putResults error of the type “concurrent or recursive” calls.
  2. If using PMPP and modules it is possible to get a structured exception when two DLLs both try to load the same table at the same time. Depending on whether fast runtime or fast compile is used in the structure, or whether a module link is set to be looped, a DLL may represent a single module shared amongst module links in the same product or individual module links.

Note: It is recommended that users apply this update if they are using PMPP.

Prophet Enterprise 2020 Q2

• There was an issue with erroneous generation of base sets when referenced directly in a run setting as part of a job. The issue generated a version for the base set, regardless of whether a base set change had been made since the last version was taken. This would then fill the database with many copies of the same base set version, which affected performance. This has been fixed so that now versions are only created when the base set data is unique when submitting a job. This is now consistent with other methods of job submission that use base sets.
• There was a performance issue that caused file system writing for certain types of results files to bypass the Windows file cache, causing some Prophet models to run noticeably slower. This has now been fixed.
• An issue which meant copying run settings from one workspace to another would, on occasion, throw an error has been fixed.
• There was an issue with the Override Results Directory names option which meant multiple run numbers could be stored in the same directory. Validation in the run setting now prevents this from occurring.
• Bugs in the Prophet Enterprise Administration Tool (PEAT), found when changing Data Storage from By Compartment to By Domain and vice versa, have now been fixed.
• There was a bug in the Prophet Grid Administrator which meant a spurious error message would appear if your web server address included a port number. This has now been fixed.
• The Prophet Enterprise runlog has been made more resilient to instances where the ArangoDB server goes down in the middle of job preparation. In this scenario, accurate start and finish times for preparation are now shown.

Prophet Results Database

• In Prophet Enterprise Release 2019 Q2, the Create a results file for download option would create a zip at a different level in the results file structure compared to previous releases. This meant that Prophet Results Database uploads using the -cri argument would not succeed. This has been fixed so that the results are once again zipped at the pre-2019 Q2 level.

Larger models benefit from the scalability achieved through the faster writing and better organisation of the results output, without relying on expensive specialised hardware.

In the 2019 Q2 release, users licensed for Flexible Results can configure Prophet Enterprise to output results to ArangoDB. ArangoDB is a NoSQL database system supporting parallel writing and file compression, and can be installed from the Prophet Enterprise installation disc. For this release ArangoDB must be installed on a Linux operating system. This is due to the greater
performance achieved on Linux. More information on ArangoDB can be found in the Prophet Enterprise Technical Guide, and on their website at https://www.arangodb.com/.

A new tab has been added to the Prophet Enterprise Administration Tool to define Arango configurations. These configurations
detail which database and server the configuration applies to, and how workers connect to the specified database when jobs are
submitted.

The data storage setup tab has also been updated to allow users to choose a results storage mechanism. Users licensed for Flexible Results can either choose to output results to the Prophet Enterprise file system as normal, or to select one of the previously defined Arango configurations. This configuration is applied at a domain level, so different domains can be set up to output to different Arango databases. This will apply to all future jobs submitted after the configuration is set, providing users the flexibility to switch between storage locations over time.

Text results and runlog information will continue to be stored under the job folder on the Prophet Enterprise file system, even if the other results are stored in Arango.

Results change
To improve the scalability of Prophet under our Flexible Results project, enhancements have been made to how results are stored in Prophet Enterprise. Flexible Results is a new licensable feature for Prophet Enterprise users, however these changes will be seen by all users including those who are not licensed for the Flexible Results feature.

To optimise these changes the names of the files stored in the results folder have been altered. It should be noted, however, that this has no effect on how results are presented in the Prophet Enterprise user interface, and the same names for the results files used previously are retained here for your convenience. Results produced by Prophet Enterprise can still be used with other Prophet applications and you can still view results produced by previous versions of Prophet in this latest release.

Looped Modules
Prophet Enterprise 2019 Q2 supports the looped modules functionality introduced in Prophet Professional 2019 Q2. Looped modules workspaces created in Prophet Professional can be uploaded to Prophet Enterprise. For more information on Looped Modules please see the Prophet Professional 2019 Q2 Bulletin.

If you are interested in seeing how Modules and Looped Modules can be used please speak to your Sales Representative or refer to the user manuals.

Advanced Vector Extensions (AVX)
Looped Modules works best in conjunction with Advanced Vector Extensions, which allows the loops of the Looped Module to be run in parallel. AVX is a set of CPU instructions that enable vectors of integers, singles or doubles to be processed simultaneously. This allows the processor to calculate a lower number instructions on the same amount of data and thus increase
performance. This will be of particular benefit to cloud users, for whom runtime directly affects costs.

Each worker process is now capable of executing different Single Instruction Multiple Data (SIMD) instructions concurrently. For example, the following four instructions might be sent to the CPU:
18 + 11
12 + 5
6 + 8
32 + 2

Traditionally, this is processed by the CPU as four separate instructions which must be undertaken one at a time (ignoring complexities around the difference between Floating Point Unit (FPU) and Arithmetic Logical Unit (ALU) processing and how these can be parallelised). With AVX, this set of instructions could be processed simultaneously, by refactoring this set of
instructions as one vector addition:

Performance is fundamentally driven by how many instructions the CPU can perform per second. By reducing the number of instructions required to process the same data, AVX can significantly increase performance.

AVX is licensable functionality, and you must also have a Prophet Enterprise licence. Your model must also make use of the Looped Modules functionality as it is the loops that are optimised to use the AVX functionality.

If you are interested in finding out more about AVX and how it can help then please speak to your Sales Representative.

Prophet Grid Administrator API
We have developed a new API for use with Prophet Grid Administrator. This API enables the use of the key functionality from Prophet Grid Administrator via an API, for example:

• Adding/deleting machines from a grid
• Adding/deleting machines from domains
• Adding/deleting/editing machine groups
• Adding/removing machines from machine groups
• Adding/removing machine groups from compartments

This API is accessible via PowerShell or .NET, and so is able to be used by Prophet Control Centre, or whichever mechanism you currently use to access our existing APIs.

Improved data transfer protocols
We are always seeking to improve the security, reliability, and speed of our software. To that end, for 2019 Q2 we have deprecated BITS and moved to WCF chunking as our data transfer mechanism.

• Upload and download of data from Prophet Enterprise servers are now handled by WCF.
• References to BITS in the Studio UI and configuration files have been removed.

Size of the batch can be specified for split by model point batch
The Model Point Batches split type has been improved. In the run setting you can now specify the number of model points that each batch contains rather than having to use the default of 10,000. This provides flexibility to tune your model performance.

Categories of finished jobs
Finished jobs were previously read-only entities. This could cause administrative issues when new categories were created in the Prophet Enterprise environment. From Release 2019 Q2 you are able to change the category of finished jobs. The rest of the entity remains read-only to protect the audit trail.

Improved security
You are now able to configure the SMTP mail server to send emails using SSL for notifications.

We have additionally made other security enhancements, which are invisible to end users.

Streamline machine entry
In previous releases you could use the Monitor Manager config file as a mechanism to add machines to your Prophet Enterprise grid. This approach is no longer supported and has been changed to simplify machine administration.

Improved error logging
Some error logging has been improved for the API, particularly with regard to serialisation errors.

Note: From this release we no longer support 32-bit compilation of Prophet code and calculations need to be performed using the 64-bit runtime architecture. Infrastructure that only supports 32-bit will not be able to be used with Prophet Release 2019 Q2.

• A bug which sometimes prevented you from accessing the runlog via the web service when using HPC has been fixed
• A bug which caused time-outs when filtering a very large list of finished jobs that were using many parameters has been fixed
• A bug which caused the scroll bar to not appear on the categories and parameters window of the job creation wizard has been fixed
• An issue which meant there could be duplicate email addresses in the Notification tab of the Job Submission dialog has been fixed
• A user interface issue with nested structures jobs which meant the same structure was duplicated in the versions dialog box has been fixed
• A bug which sometimes caused the runlog to not be indexed after resubmitting a failed job has been fixed
• A bug with the Allow Concurrent Model Point Batches check box in a product which meant that the correct value could be lost on creating a new version of a workspace has been fixed
• Having an expired password on the target environment when performing a promote caused an error. This has been fixed and users are now correctly prompted to reset their password.
• An issue with reconnection via the API after a session is lost has been fixed
• Some issues with errors produced during a database upgrade have been fixed