TigerGraph Quick Start

The purpose of this guide is to help you quickly gain familiarity with the core concepts behind Graph and how to navigate a TigerGraph Instance through Graph Studio.

Accessing the Graph

The "Social Graph" Test Drive is built to demonstrate how we can use graph to uncover non-obvious connections between individuals.

The Schema

The link above will land you on the Design Schema page for the Social Graph. You'll see a bunch of different colored circles with icons in them with lines between them. Those circles are our nodes and the lines are our edges.
Hovering over each node or edge with your mouse will reveal its attributes. You can think of each node type as a table in a traditional database and each edge as a pre-computed join. Each individual node of a particular node type represents one "row" of data. The power of Graph comes from our ability to quickly traverse these edges to find interesting connections between our nodes.

Click around in the Schema and see if what insights you think we can gleam off the data in the graph.

Data Mapping

On the left side of the interface, click the next tab under Design Schema, Map Data To Graph. This tab shows us how each node and edge makes use of our loaded data. The Paper Icons represent our different data sources. You can see that these sources are just .csv files. For demo purposes, viewing the source data files is disabled, but you can get an idea of what each data source looks like by clicking on one of the dotted line emanating from it.

On the right half of your screen you'll now see at least two tables. Select the dotted line between bank_account.csv and the bank_account_own edge. The table on the left represents our data source as we can see from the "file: bank_account.csv" title. The right hand table represents our edge and more specifically the source vertex, target vertex, and any attributes associated with the edge. In this case we only have two vertices to populate.

The teal arrows represent the mapping between our data source and edge type. We can see that the bank_account_own edge connects a citizen vertex to a bank_account vertex by looking at the left window with our data mapping. And by looking at our teal arrows, we can see that we are mapping the OpenAccountPersonID to our "citizen" source vertex and the CardNumber field to the "bank_account" target vertex.

Data Loading

Next we'll take a look at the Load Data tab on the left. This tab shows us our data mapping again, but also displays some statistics about our loaded data and the button to actually start the data loading process. Additionally, you can click on any Data Source and view its loading logs with the clipboard icon at the top.

As you load in data, you'll be able to see the line graph on the lower right respond to the loaded data. Unfortunately, we've already loaded the data in our demo so you'll just have to set up your own graph to witness the awesome sight of billions of nodes being loaded in real-time.

Explore Graph

You'll never guess what the Explore Graph page lets us do. Alright, maybe you did guess it. Here we can explore connections between specific nodes in our graph without having to write any queries. You'll see that this section is actually broken up into five sub-sections denoted by the icons on the left side of the window. I won't go into too much detail on these sub-sections here as you can learn all about them in our Graph Studio Intro Lessons.

We'll poke around in the Search Vertices sub-section first. This is the magnifying glass icon. The Vertex Type dropdown lets us select the vertex type that we would like to search for, and the Vertex ID field lets us specify the vertex. This is great for if you have a specific point of data that you're looking to investigate. Alternatively, you can just use the Pick Vertices button to select the entered number of vertices of all types. This can be great for getting a general feel for your data.

Write Queries

Here's where we can really see TigerGraph stretch its legs. Queries allow us to explore the graph in ways that just aren't possible with traditional databases. You'll need your own TigerGraph instance to write your own queries, but don't worry, you can get running for free with Tiger Graph Cloud. We have one query pre-installed called connection_mining. This query will determine the strength in the connection between two of our citizen nodes by analyzing the connections between those initial citizens. The query will traverse all connections associated to each individual and return paths connecting the two citizens.

The query requires that we insert two starting vertexes and a K integer. We can use 7747635 and 753364 as our starting citizens but really you can type any number in here to explore random citizens. The K integer is the number of people associated to each starting citizen that we're going to consider in our strength calculation, 200 is a good number here but play around.

Your results will appear below the query and there are two different ways to view them. If you query returns vertices (ours does) then the query result will show a graph view of the vertices and any returned connections. You can also display the JSON Result of the query by selecting the icon under the multiple dots. This JSON Result is identical to what you would get returned via API call when calling your query through the TigerGraph REST API.

Why not get TigerGraph certified today?