From zero to dashboards
A step-by-step tutorial guiding you from creating your SlicingDice account to building reports and dashboards in the Data Exploration & Visualization Module.
Introduction and Prerequisites
This tutorial is made to quickly set up your database and use our modules. You'll be able to access the Control Panel to work through your whole data pipeline, from raw data to analytics.
To be thorough this guide will assume:
- You know almost nothing about SlicingDice itself.
- You don't have a SlicingDice account yet.
- You want to load an existing dataset or insert data being generated now (or both).
- You want other users to access this data (or part of it), to explore and visualize it.
- You want this data to be used to make complex analyses (queries).
- You want to create charts and dashboards that can be accessed by other users.
- You want to run advanced analytics on this data, such as machine learning.
In order to implement SlicingDice and get from zero to data analysis, dashboard creation, and run machine learning jobs, you'll only need:
- General knowledge about database concepts.
- General knowledge on running queries using SQL.
- Access to the database or dataset you want to load.
- General knowledge about the content of the selected database or dataset in order to model your data into the SlicingDice’s database.
1. Register to SlicingDice
Register at our website to access our
control panel using only an email and password.
SlicingDice offers free testing, no credit card needed.
2. Download our dataset
If you want to be practical, we recommend to you to download the example dataset that we'll be using in this tutorial.
3. Creating a database
There are two ways of creating a database in SlicingDice: Using the API or your Control Panel. This guide will show how to create a database using your control panel (please access our API documentation if you want to create a database using API).
To create a new database using the Control Panel, access the Physical databases pages inside the Data Warehouse Module and click on Create New Database.
A 3-step wizard will appear.
The Step 1 is where you'll need to provide basic database information, like name, type and pricing model.
|Database Name||Yes||The name of your database. Can be edited at any time.|
|Database Labels/Tabs||No||Possible labels/tags you might want to associate to a database, in order to organize it. Can be edited at any time.|
|Database Description||No||The description of your database. Can be edited at any time.|
|Cloud Provider/Region||No||The cloud provider and its supported region that will be used to store and query your data. Cannot be edited after creation.|
|Database Type||Yes||Can either be Test or Production. It's only possible to change a Test database to Production, not the other way around.|
|Pricing Model||Yes||The pricing model you want to use for your database. Can be changed at any time.|
|Insertion Load||Yes||The insertion throughput you want your database to have. Can be changed at any time.|
|Query Load||Yes||The query throughput you want your database to have. Can be changed at any time.|
These parameters in the first step are the only ones needed to create a database.
Notice that in the following steps you'll be able to define other parameters such as your schema definition, IP whitelist and Custom Database Keys creation. You'll be able to define and edit these steps later, check how to do it accessing the modeling a database schema and securing a database guides.
This way you can ignore the steps 2, 3 and 4 if you want by clicking the Save & Continue button until completion.
SlicingDice allows loading data from any source, which requires no modelling and handles different data structures seamlessly. This allows you to insert and leverage data without setting up an ETL process, saving you time and money.
In this section, you'll insert data using our built-in Data Loading & Transformation Module in
the Control Panel.
To load your data, you'll need to go through 3 steps:
- Connect to a data source
- Create a loading job
- Load the data
Setting up a data source will save all the connection information (such as location and credentials) needed for its access by SlicingDice.
Below we explain how to add a new data source.
In the Control Panel, go to the Data Sources page inside the Data Loading & Transformation Module and click on Create New Data Source to start connecting your external data source.
The first step is to configure your data source identification as you can see below.
Here you'll define your data source name, description and possible labels. When ready, click on Save & Continue.
Now you'll define the appropriate data source credentials. Each data source type has its own credentials.
Common credentials are
region and so on.
In this example, we'll add the HTTP address of our dataset.
You can test the connection by clicking on Test Connection. If everything goes ok, you'll see a Connection Success message. Click on Save & Continue to proceed.
If you need to connect using a SSH Tunnel, take a look at the Creating an SSH Key connection guide.
Finally check your data source configuration and click on Submit. Now you'll be able to
find your created data source at the data sources list.
That's it! Now you can use this data source to create a loading job that will connect to this source and load your data.
If you need to whitelist the IPs which SlicingDice will use to connect to your source, use the following ones shown below.
A loading job is a data loader that will connect to a given data source and import external data into
Loading jobs have three different types:
- One-Time Manual Load: Used if you want to load data from a source just once.
- Manual Incremental Load: When executed, loads only the data that was inserted since the last time you executed this process. Requires a timestamp field.
- Automatic/Scheduled: Will load data indefinitely and according to a defined frequency. Requires a timestamp field.
To create a loading job, go to Control Panel, access the Loading Jobs section on the Data Loading & Transformation Module and click on Create New Loading Job.
The first step is to configure your loading job identification as you can see below.
Here you'll define your loading job name, description, the registered data source to connect and the
job type. For this example, we'll select the One-time Manual load, as our dataset is static.
When ready, click on Save & Continue.
Now you'll define in which of your databases and dimensions the data will be loaded.
When everything is set, click on Save & Continue.
After this, you'll map your source columns and match them with the SlicingDice columns and their supported data types.
For it, the first step is to detect your source schema, selecting the data you want to retrieve. You can do it choosing the Access mode - query or table - where you can define a SQL query or just inform the name of the table to retrieve data.
If the name of the column or file has more than one word or has some separator, you'll need to use
In this tutorial, we'll import the whole dataset, and we need to pass the name of the file, which is
drinks.csv, so you need to insert it within brackets, like the example
After detecting your schema, the selected source columns will appear on the Attribute and Event Columns mapping section. Map your source columns to SlicingDice existing columns or create new ones if needed. On this example, you'll create new columns. Click on Create Column.
Insert the column name, the column type and the storage. The following table shows how to map our dataset. Be in mind that all storage types are latest-value
|Data Source Column (type)||Destination Column (type)|
|wine_servings (int)||wine-servings (Integer)|
|country (varchar)||entity-id (Unique-id)|
|spirit_servings (int)||spirit-servings (Integer)|
|total_litres_of_pure_alcohol (float)||total-litres (Decimal)|
|beer_servings (int)||beer-servings (int)|
After mapping, click on Save & Continue.
Mapping an Entity ID is mandatory!
On SlicingDice an Entity ID column behaves similarly to a primary key column in a database with unique values stored on it, so any data you insert will always be associated with a single identifier. This is why you MUST define one of your columns from your sources to be your Entity ID column, otherwise you'll receive an error while creating your loading job.
Now check your loading job configuration and click on Submit. Now you'll be able to find
your created loading job at the loading jobs list.
That's it! The next step is to execute your loading job.
Click on the Play icon, and your dataset will start to load into SlicingDice.
The Loading Job execution time can vary by the size of your dataset and the chosen Rows per request per second configured at step 3. When the "Status" parameter modifies to "Completed", you can start querying your data.
Automatic Loading Job
If you set your Loading Job as "Automatic/Scheduled", you won't find the Play icon. This type of Loading Job will execute automatically, according to the frequency.
Permission groups allows you to define fine-grained authorizations for users to interact with the databases, inserting and querying data, using SQL or API, and also get access to the Control Panel and its modules.
You can create a permission group using the Control Panel by accessing the Permission Groups section on the Identity and Access menu on Account Management where you'll be able to see all the permission groups available in your SlicingDice account. Now you can click on Create New Role and Policy to start the Permission Group Creation Wizard.
First you'll need to define a valid name and description (optional) for this permission group.
After that click on Save & Continue and define which access permissions this permission group will contain, optional labels and also set specific week days and time that users associated with this permission group can access. You can check the available permissions on the permissions list.
Now, click on Save & Continue and check all the information you've set to this permission group. If everything is fine, you can finish this permission group creation, making it available in your permission's groups list.
SlicingDice allows you to create users that will be able to access their accounts and use all the capabilities that they have permission to use, according to the Permission Group which they belong to.
You can create an user using the Control Panel accessing the Users page on the Identity and Access menu on Account Management where you'll be able to see all the users currently associated with your SlicingDice account. Now you can click on Create New User to start the User Creation Wizard.
First you'll need to define a valid e-mail, optional labels, the language and you can optionally set a password for this user that is being created. If you don't set this password, SlicingDice will send a random one to the user's e-mail.
After that click on Save & Continue and define which Permission Group this user will belong to.
Now, click on Save & Continue and check all the information you've set to this user. If everything is fine you can finish this user creation, making it available in your user's list.
SlicingDice has powerful built-in Data Exploration & Visualization Module that allow you to easily create and share beautiful and useful charts, dashboards and presentations, using data stored in SlicingDice.
In the Data Exploration & Visualization Module, you can create charts using your data stored on SlicingDice.
To create a chart, you need to create a dataset. Watch this quick video or read the text.
In the Reporting & Dashboards Tool page inside the Data Exploration & Visualization Module,
Data sources and click on the database name you
Click on Browse Datasets, then click on Write SQL query option.
You can use the following SQL code to select all your data from your
SELECT * FROM default;
Insert the SQL code and click on Run.
After that, click on the Done button, choose a name for your database, then click on Rename.
All set, click on Done button again. That's it!
After the dataset creation, you are now able to create a chart.
Charts show quantitative measures grouped by qualitative dimensions. Different chart types work with different numbers of measures and dimensions. Watch this quick video or read the text.
- Click on the database, then on the dataset you just created
- Click on Create Saved Chart
- Choose a chart type
- Choose the dimension and a measure of the chart
- Click on Done
- Click on the vertical three dot option and click on Save
- Select the name of your chart, then click on OK.
Dashboards are a fundamental part of the Data Exploration & Visualization Module.
You use them to group multiple charts in the same screen and create interactions between them.
This well-structured and easy to analyse dashboard can be shared with your team with a simple URL. To create a new dashboard, watch the following video or follow the steps below.
In the Reporting & Dashboards Tool page inside the Data Exploration & Visualization Module,
Dashboards on the main menu.
Click on the plus signal (+) located at lower right of the page. You'll see a blank canvas grid.
Insert image, charts, text and more clicking on the Insert button.
Click on the vertical three dot option and click on Save.
Select the name of your dashboard, then click on OK.
The Data Exploration & Visualization Module gives you the option to build presentations, which allow your charts
updated in real time.
Watch the following video or follow the steps below.
In the Data Exploration & Visualization Module, click on Presentations on main menu. Click on
signal (+) located at lower right of the page. You can now insert content to your presentation.
Click on the Insert button to add charts, images, text or a new slide. When everything is set,
click on vertical three dot option and click on Save.
Select the name of your presentation, then click on OK.