Explain Tableau server Architecture
Information about the Tableau architecture is critical in our attempt to learn about the new BI method. You would be able to get a clear technical understanding of the device by knowing what is under the hood. In this article, let us examine the design of the tableau and comprehend in depth it’s working. Tableau has a highly scalable client-server architecture n-tier that serves mobile clients, web clients, and software installed on the desktop. Moreover, the Tableau Desktop is the authoring and publishing application that uses Tableau Server to create shared views.
The Tableau Server is a business-class analytics application. It is to level thousands of customers up. Moreover, it presents powerful analytics based on mobile and browser. Then it works among the data strategy and security protocols presented by a company.
Components of Tableau server architecture
Let us research the different components of the Architecture Tableau.
Application Software
Tableau Architecture’s primary component is the data sources it can link.
Tableau can link to multiple sources of the data. You can store such sources of data on-premise or remotely. It can all simultaneously connect to a database; excel file, and a web application. Tableau can connect heterogeneous ambient data. It can use the data from multiple sources of data. It can also make the connection between different types of sources of data.
Data Couplers
The Data Connectors provide an interface for connecting Tableau Data Server to external data sources.
Tableau has an ODBC / SQL connector built into it. Without using their native connector this ODBC Connector will connect to any database. Tableau provides an option for selecting both live and extracts data. One can easily switch between extracted and live data, depending on the use.
Live Connection or Real Time Data:
Tableau can connect directly to data in real time by linking directly to the external database. This uses existing database system technology by sending out complex MDX (Multidimensional Expressions) and SQL statements. This feature will connect Tableau to the live data instead of importing the data. Its goodness is an organization’s investment in a strong and configured database system. In many enterprises, you can regularly update the database size and it is enormous. In those instances, by connecting to the live data, Tableau acts as a front-end visualization tool.
Information extracted or in-memory:
Tableau has the option of collecting data from external sources. You may create a local copy in tableau-extract file format. It can take a single click to retrieve millions of records in Tableau data engine. Tableau’s data engine uses storage to store and process data, such as RAM, ROM, and cache memory. Tableau will remove only a few records from a huge dataset using filters. This enhances performance, especially when working on large data sets. Extracted or in-memory data enables users to view offline data without connecting to the source of the data.
Tableau Server components
In a Tableau server the various components present are as follows.
● Server Service
● VizQL Database Configuration
Application Software
- A) Database applications:
The application server provides the authentications and permissions. It handles the Web and Mobile Interfaces administration and permission. Recording each session Id on Tableau Server ensures security. The Administrator can configure the server session’s default timeout.
- B) Database VizQL:
You can use VizQL Server to convert the data source queries into visualizations. Once you forward the client request to the VizQL process, it sends the query directly to the source of the data and retrieves information in image form. The user can present with this image or visualization. Tableau server builds a visualization cache to reduce the load time. Many users who have the permission to access the visualization are able to share the cache.
- C) Server data:
Data server uses external data sources to handle and store data. It is a central system for the management of data. Moreover, provides the metadata management requirements, data security, data storage, data connection, and driver. Then it stores the relevant data set information including metadata, computed fields, sets, classes, and parameters. Thus, the source of the data could also collect data and make live links to external data sources.
Gateway
The gateway channels user requests into components of the Tableau. Once you receive a request by the client, you can forward it for processing to an external load balancer. The gateway acts as a process distributor to different components. Gateway also acts as a load balancer in case of absence of external load balancer. For single server setup, one primary server or gateway handles all the processes. One physical machine serves as the main server for several server setups, while others use it as the worker servers. In Tableau Server, setting only one computer as the primary server.
Clients
The Tableau Server accesses and edits dashboards and visualizations using different clients. Clients are Tableau Desktop, Smartphone apps and web browsers.
Tableau Server is essentially a communication tool that shares end-user or client data connections and visualizations. So, now that you have learned about each part operating in a Tableau server. Let us understand how all of those components work together. You must club the components of the application into layers or strata for this. Therefore, in the Tableau Server, you have five layers or sections; consumer data, data connectors, main components, gateways and clients.
Tableau Architecture
The customer data layer contains all kinds of data sources available to a Tableau user, such as data warehouses, data marts, flat files, and relational databases, multidimensional cubes.
Next are the layers of data connectors consisting of a data engine, registry, SQL connector, and MDX Connector. Such components directly communicate with the sources of data. The Data Engine processes user-requested data and determines the type of data, defines whether it is a metric or a component, and generates TDEs (data extracts). Engine runs a SQL Connector at the data context, which generates a SQL query for all user requests and interacts with the data sources. Above all, the SQL Connector deals with data marts and flat files. The MDX Connector likewise deals with the multi-dimensional cubes.
The next layer includes all the main components, the data server that controls and tracks the operation of the Data Connector layer components. This contains a VizQL Database, as well as an Application Server. To access the visualization, the application server takes all user requests that come from Tableau Desktop, Smartphone or browser. It handles the requests, identifies the type of request, checks user authorization, and accordingly grants access to them. The VizQL Server is a proprietary Tableau component, where VizQL stands for the language Visualization Query. It works behind the visualization logic of Tableau, and creates visualization as per your dashboard instructions.
As you have already learned about the gateway, it acts as a Tableau Server gatekeeper, and any request or query that the client sends first hits the gateway or load balancer. A gateway is nothing more than a primary server that receives the queries and redirects them to an appropriate secondary server, known as a worker server.
Another way of organizing the architecture for the Tableau Server is through a tier-based model. Then divide architectural components in such a model into three parts, i.e. a consumer tier, storage tier, and a management tier
Conclusion
I hope you reach to a conclusion about Tableau architecture. You can learn more through tableau server online training.
