FAQ

Find the most frequently asked questions and answers here

I am now using trigger files, how can I deal with deltas now?

In Snowflake the data becomes available in views, these views have the upload date/time of each record. This allows you to select deltas.

Can offices also create their own views?

The data platform is provided by ANVA. Therefore, it is not possible to create your own views.

What does the file look like?

We have a sample file in Excel.

Can I run multiple queries at the same time?

It is possible to run multiple queries simultaneously. If sql-api is used, a request can contain by default only 1 sql statement, if more statements are used in a request it must be explicitly indicated in the API call.

How can I maintain the best performance when running queries on the data platform?

We do not recommend working with "select * " in selections. This is for two reasons. First, it is detrimental to performance. And secondly, this construction will mean that if we make a change in the table the processing of the query will produce errors. And select with specific columns, only the columns that are needed, will continue to work and be faster.

How can I create and run queries on Snowflake?

Snowflake uses standard SQL supplemented by snowflake proprietary functions. Snowflake prefers to work with capital letters, so table names, column names that are not that are not capitalized should be double-quoted.

Some queries (yet to be defined) may take longer

Performance is strongly influenced by the data that is requested. We recommend only requesting the data (columns) that are actually needed for further processing. All additional data is detrimental to performance.

I use large queries

Snowflake is capable of processing large queries quickly, the data is then persisted so that the requesting application (e.g. SQL-SERVER) can read this data. However, this data remains available for a limited time for security reasons. This time is set by Snowflake to 6 hours and cannot be adjusted.

The data of the query must be read within 6 hours, after these 6 hours an authentication error will occur.

You need to change something in your registry, how, what?

After installing the ODBC connector, the credentials must be added in the registry. The interface for configuring the ODBC connector does not provide for adding the keypair credentials.

I am now using the 32-bit version of the ODBC connector

For the 32-bit Veryant ODBC link, separate EFDs were created in the past for some tables where the record length exceeded the supported length.

The tables in question are NAWBES, NAWHIS, POLBES and POLHIS. Four separate EFDs were created for each table for this purpose (e.g., NAWBES_DEEL1 to NAWBES_DEEL4),

In Snowflake these tables are full length. So you will not find the PART tables in Snowflake.

What does it mean that the variable part is broken down?

From ANVA, all labels come in 1 long string. We have broken down the variable parts and made them available in separate views. This makes it easier to read into your systems.

These views display the labels from the varpart column in a useful way. These views have the name of the table with VARDEEL appended, so NAWBESVARDEEL or POLBESVARDEEL.

However, it is not recommended to retrieve the labels of all relations or policies with 1 query as these views contain a bit more complexity and thus require more time to collect the data. Especially the historical tables will be large so retrieving all vardeel labels can take a long time.

To address this performance issue, we developed an additional set of views for the variable sections. These views do not perform joins with other tables; the data is displayed unfiltered and unformatted. This results in a significant performance improvement.

A disadvantage is that for some labels with more complex notation, the value is not displayed as nicely. These views have the name of the varpart views with the suffix _COMPACT, such as NAWBESVARDEEL_COMPACT or POLBESVARDEEL_COMPACT.

During extracting for the data platform, in the variable fields, for example NAWBES or POLBES, the fieldseperator that separates the labels is filled in with a |-sign.This means that it is not wise to use that |-sign in values entered in labels.In the VARDEEL views, we split the labels using the |-sign and we then look for a |-sign with 7 numbers (labels and length of the value) behind it. If there is a |-sign in a label value the text after the |-sign is not shown in the VARDEEL views.

What do date field adjustments mean?

The raw data from ANVA contains a number of inaccuracies. For example, it sometimes contains incorrect data units, such as Feb. 31. We have adjusted this to correct dates. Also, for example, we have made date fields real date fields.

What do prefixes mean?

A prefix is added to column names to quickly see which table it comes from. For example, the prefix "AGB_" is added to columns from the table AGBES. Then, for example, you get AGB_AGENTNR instead of AGENTNR.

How does my data get into the data platform?

To bring the ANVA data from your system to the data platform, we use the tool ANVA extract. With ANVA extract, the data is dumped into CSV files that are then transferred to our AWS environment via a secure connection. The CSV files are deleted from your system after the transfer to free up space.

How can I see what data has been added to the data platform?

All data has timestamps, this allows you to see when the data was added to the data platform. These can be found in the table meta_load_times.

Is the var part labeled?

Fixed fields are labeled though. So numbers are numbers, amounts are amounts, etc.

Where can I register if updates need to be done by Snowflake ODBC connector?

You can subscribe to Snowflake to be notified regarding updates to your Snowflake ODBC connector. To do so, go to: https://docs.snowflake.com/en/release-notes.html

Timestamp for changes/deltas to be insightful. How can you create a view on this?

The available views have the current state of the data. A timestamp is present in the ODBC views to identify changes or new records. The timestamp represents the date time that the data was loaded into the data platform and not when the change was made in ANVA.

In the variable tables, can I see the code in addition to the description of the tables?

In the ODBC schema in the data platform, we have done an extension in the variable tables. These tables also have LABEL_DISPLAY_WIJZE_CODE in addition to the description. This saves tremendously in storage and you can create your own translation in your environment based on the code.

How can I tell if the extract was loaded successfully?

When an extract is loaded, data about this extract is recorded in a table. The data from this table is also made available through a view named META_LOAD_TIMES, both in the ODS and in the ODBC schema.

In this view you can see the LOAD_TYPE which indicates what type of extract was performed. The LOAD_TIME indicates the time when the extract was loaded into Snowflake. Each table appears twice since data first enters a STAGING schema and then is passed into the ODS schema. The data for the ODS schema(SCHEMA_NAME = 'ODS') will be most relevant since it indicates the LOAD_TIME when the data actually became available. The NUMBER_RECORDS column indicates the number of records in the table after processing, the DIFFERENCE column indicates how the number of records has grown or decreased since the previous processing.

What is the cost of using the ANVA Data Platform?

Each customer receives a free to spend number of data platform credits. That number depends on the number of policies and contracts you manage with ANVA. If you use more credits, you will be charged for the number of extra credits used. Do you have questions about the use and costs of the data platform? Please contact your account manager.

Where can I find the terms and conditions that apply to the proper use of all ANVA solutions as a Developer?

The correct way to use the ANVA solutions offered is described in our Terms of Use for you as a customer and developer. There you will also find an explanation of our fair use policy for the ANVA Data Platform.

May I use your data solutions for AI and training my own language model?

The use of our Data Solutions for AI purposes and/or for training a language model using the infrastructure and techniques we provide is prohibited. However, you are free to transfer the data you own from our solutions to a technical environment that we do not manage. This allows you to work with AI outside of our solutions. The correct way to use the ANVA solutions offered is described in our Terms of Use for you as a customer and developer. There you will also find an explanation of our fair use policy for the ANVA Data Platform.