How Graph reduces data locked-in risk?

thumbnail

on May 26, 2020 • Back to Blog index

Whenever a new enterprise IT solution hits the market, the industry is a bit skeptical about the risk of creating data in another system. We do believe that Product Lifecycle Management is mainly about organizing intellectual property into nodes of information related by semantic relationships. In other words we store bits of information relevant to your business linked by relationships that mean something to you. These information have to be stored on the long run but also need to be available quickly for users when they need it.

This is a big decision to put some of the most valuable data in a new system. How do we, Ganister, mitigate the risk for you.

What risk?

Lets quickly define the actual risk levels:

Capability to retrieve your data

This is purely technical, are you allowed to get access to your data.

Capability to read the content once retrieved

This is about being able to read what is in the data, retrieving the list of tables that contains your data

Capability to understand the content once retrieved

This last one is a bit tricky. A PLM system can be a bit complex. If what you get is a list of tables representing items and relationships, how will you retrieve the whole knowledge structure easily?

The graph database again

One feature that we really appreciate even though it may not be perceived as a killer feature at first, is the semantic used in the database. A graph database is a big set of nodes and relationships. All your data can be queried using the actual semantic of your business nothing specific to Ganister.

If you needed to retrieve your full set of data, Graph databases typically allow to export the whole database as a big Json Object. What's a Json Object? Let's take a small dataset as an example and see how it gets exported.

Let's create a small dataset with two nodes and one relationship

CREATE (a:Company {name:'WorldCorp'})
, (b:Employee {name:'Jim', age:42})
, (b)-[:WORKSAT {since: 1993}]->(a)

Here is the Neo4j playground console:


The export of such data, would look like this

{
    "backup": [
        {
            "type": "node",
            "id": "0",
            "labels": ["Company"],
            "properties": {"name": "WorldCorp"}
        },
        {
            "type": "node",
            "id": "1",
            "labels": ["Employee"],
            "properties": {"name": "Jim","age": 42}
        },
        {
            "id": "0",
            "type": "relationship",
            "label": "WORKSAT",
            "properties": {"since": 1993},
            "start": {
                "id": "1",
                "labels": ["Employee"]
            },
            "end": {
                "id": "0",
                "labels": ["Company"]
            }
        }
    ]
}

It may still look technical but it actually is very simple to parse for any IT person. And even a non it person can read that there are two nodes and one relationship.

Cloud-first delivery

Obviously the other reason why your data is safe with us is because of the cloud distribution we prefer for our customer. We constantly backup your database and monitor access to make sure only your Ganister application and your company interact directly with your data.

Let's discuss about data safety with Ganister: