Differences between the Community and Enterprise Editions of Datafari

The conceptual difference between the community and enterprise editions relate to the following themes: security, big data, exploitation and relevance. This means that whenever we provide advanced functionalities for one these axis, we tend to keep it for the enterprise edition rather than the community one, in order to maintain a strong interest for potential customers to get the Enterprise Edition. We do not put any particular constraints on the community edition, therefore it remains possible to implement by yourself any functionality that we do not provide outside of the Enterprise Edition, and there are high chances that we will not accept the work as a contribution that would be integrated to the community edition (as it would be conflicting with our own functionality)..

If we take these axis and drill down, this gives us, in terms of functionalities only available in our Enterprise Edition:

1. Security

   1. Graphical AD configuration: the possibility from the Admin UI to setup the connection to your AD;

   2. Multiple AD OU: the technical capacity for Datafari to manage several AD organisation units;

   3. SSO: Integration with Kerberos and SAMLv2;

   4. Caching mechanism for the users roles: Caching user connection to lower the load.

2. Big Data

   1. Multiservers clusterisation: preconfigured environments for a multiserver environments (including the SolrCloud capability);

   2. Full compenentisation via Docker containers: all the components of Datafari (Cassandra, Postgre, MCF, Solr ...) are packaged within their own container, allowing for a "micro-service" approach;

   3. Enhanced Data Extraction Server: Preconfigured and faster indexation phase.

3. Exploitation

   1. Backup scripts: automatically backup your MCF configuration, your Solr index. Scripts and documentation to reload the backups.

   2. Advanced Monitoring: Configuration of Glances, logs downloadable per component from the admin UI, cron job watching the processes and restarting them automatically

   3. Monitoring Dashboards: Dashboards (based on Apache Zeppelin) to analyse the technical components, and to quickly find file indexations problems (such as missing files reported by an employee)

4. Relevance

   1. Automatic Relevancy Algorithm optimizer: Graphical tool that automatically calculates a local optimal solution for a given number of fields as part of the relevancy algorithm (leveraging the Golden Query Retriever functionality).

   2. Golden Query retriever: Graphical tool in the search results, to compile the golden queries consisting of queries/relevant documents.

   3. Machine Learning for documents real time ranking:

   4. Query time concept extractor:

   5. Indexing time concept extractor:

   6. On the fly semantic annotator: Able to connect asynchronously to 3rd party tools, such as AI tools from Google or Amazon, to extract/analyse documents before indexation. For instance, images of cars would be enriched with metadata giving the brand and color of the car; videos would get back the transcription text.

This separation means that at France Labs, by default (but there can be exceptions), work done on these axis would be put only on the Enterprise Edition, not the Community Edition. This also means that work done on other axis is by default for both versions.

Our vision is that the Community Edition should provide a smooth experience for users: easy installation, easy configuration (via the simplified connectors), and fast speed of search. There should be no particular difference on that point between the Community and Enterprise Edition. 

As for connectors, it should be decided on a case by case.

We compiled below a comparison matrix of both versions, which we try to keep as exhaustive as possible:

The conceptual difference between the community and enterprise editions relate to the following themes: security, big data, exploitation and relevance. This means that whenever we provide advanced functionalities for one these axis, we tend to keep it for the enterprise edition rather than the community one, in order to maintain a strong interest for potential customers to get the Enterprise Edition. We do not put any particular constraints on the community edition, therefore it remains possible to implement by yourself any functionality that we do not provide outside of the Enterprise Edition, and there are high chances that we will not accept the work as a contribution that would be integrated to the community edition (as it would be conflicting with our own functionality)..

If we take these axis and drill down, this gives us, in terms of functionalities only available in our Enterprise Edition:

1. Security

   1. Graphical AD configuration: the possibility from the Admin UI to setup the connection to your AD;

   2. Multiple AD OU: the technical capacity for Datafari to manage several AD organisation units;

   3. SSO: Integration with Kerberos and SAMLv2;

   4. Caching mechanism for the users roles: Caching user connection to lower the load.

2. Big Data

   1. Multiservers clusterisation: preconfigured environments for a multiserver environments (including the SolrCloud capability);

   2. Full compenentisation via Docker containers: all the components of Datafari (Cassandra, Postgre, MCF, Solr ...) are packaged within their own container, allowing for a "micro-service" approach;

   3. Enhanced Data Extraction Server: Preconfigured and faster indexation phase.

3. Exploitation

   1. Backup scripts: automatically backup your MCF configuration, your Solr index. Scripts and documentation to reload the backups.

   2. Advanced Monitoring: Configuration of Glances, logs downloadable per component from the admin UI, cron job watching the processes and restarting them automatically

   3. Monitoring Dashboards: Dashboards (based on Apache Zeppelin) to analyse the technical components, and to quickly find file indexations problems (such as missing files reported by an employee)

4. Relevance

   1. Automatic Relevancy Algorithm optimizer: Graphical tool that automatically calculates a local optimal solution for a given number of fields as part of the relevancy algorithm (leveraging the Golden Query Retriever functionality).

   2. Golden Query retriever: Graphical tool in the search results, to compile the golden queries consisting of queries/relevant documents.

   3. Machine Learning for documents real time ranking:

   4. Query time concept extractor:

   5. Indexing time concept extractor:

   6. On the fly semantic annotator: Able to connect asynchronously to 3rd party tools, such as AI tools from Google or Amazon, to extract/analyse documents before indexation. For instance, images of cars would be enriched with metadata giving the brand and color of the car; videos would get back the transcription text.

This separation means that at France Labs, by default (but there can be exceptions), work done on these axis would be put only on the Enterprise Edition, not the Community Edition. This also means that work done on other axis is by default for both versions.

Our vision is that the Community Edition should provide a smooth experience for users: easy installation, easy configuration (via the simplified connectors), and fast speed of search. There should be no particular difference on that point between the Community and Enterprise Edition. 

As for connectors, it should be decided on a case by case.

We compiled below a comparison matrix of both versions, which we try to keep as exhaustive as possible:

The conceptual difference between the community and enterprise editions relate to the following themes: security, big data, exploitation and relevance. This means that whenever we provide advanced functionalities for one these axis, we tend to keep it for the enterprise edition rather than the community one, in order to maintain a strong interest for potential customers to get the Enterprise Edition. We do not put any particular constraints on the community edition, therefore it remains possible to implement by yourself any functionality that we do not provide outside of the Enterprise Edition, and there are high chances that we will not accept the work as a contribution that would be integrated to the community edition (as it would be conflicting with our own functionality)..

If we take these axis and drill down, this gives us, in terms of functionalities only available in our Enterprise Edition:

1. Security

   1. Graphical AD configuration: the possibility from the Admin UI to setup the connection to your AD;

   2. Multiple AD OU: the technical capacity for Datafari to manage several AD organisation units;

   3. SSO: Integration with Kerberos and SAMLv2;

   4. Caching mechanism for the users roles: Caching user connection to lower the load.

2. Big Data

   1. Multiservers clusterisation: preconfigured environments for a multiserver environments (including the SolrCloud capability);

   2. Full compenentisation via Docker containers: all the components of Datafari (Cassandra, Postgre, MCF, Solr ...) are packaged within their own container, allowing for a "micro-service" approach.

   3. Enhanced Data Extraction Server: Preconfigured and faster indexation phase.

   4. Dedicated Zookeeper for the Crawling Connectors Framework (MCF): allows for better performance since the Zookeeper used by Solr is freed from answering to the MCF frequent calls.

3. Exploitation

   1. Backup scripts: automatically backup your MCF configuration, your Solr index. Scripts and documentation to reload the backups.

   2. Advanced Monitoring: Configuration of Glances, logs downloadable per component from the admin UI, cron job watching the processes and restarting them automatically

   3. Monitoring Dashboards: ELK based dashboards to analyse the technical components, and to quickly find file indexations problems (such as missing files reported by an employee)

4. Relevance

   1. Automatic Relevancy Algorithm optimizer: Graphical tool that automatically calculates a local optimal solution for a given number of fields as part of the relevancy algorithm (leveraging the Golden Query Retriever functionality).

   2. Golden Query retriever: Graphical tool in the search results, to compile the golden queries consisting of queries/relevant documents.

   3. Machine Learning for documents real time ranking:

   4. Query time concept extractor:

   5. Indexing time concept extractor:

   6. On the fly semantic annotator: Able to connect asynchronously to 3rd party tools, such as AI tools from Google or Amazon, to extract/analyse documents before indexation. For instance, images of cars would be enriched with metadata giving the brand and color of the car; videos would get back the transcription text.

This separation means that at France Labs, by default (but there can be exceptions), work done on these axis should be put only on the Enterprise Edition, not the Community Edition. This also means that work done on other axis is by default for both versions.

Our vision is that the Community Edition should provide a smooth experience for users: easy installation, easy configuration (via the simplified connectors), and fast speed of search. There should be no particular difference on that point between the Community and Enterprise Edition. 

As for connectors, it should be decided on a case by case.

We compiled below a comparison matrix of both versions, which we try to keep as exhaustive as possible: