There are several ways one can deploy Jet Analytics. This article will detail supported options, along with some general guidance and advice for a successful deployment.
Choosing a deployment model can be difficult as there are many factors to consider. These choices are usually best determined by your organization working in conjunction with a technical consultant.
There are several components that comprise a deployment.
1. Jet Data Manager (JDM)
The Jet Data Manager is a .NET application that resides on a Windows Server. The JDM is the back-end development tool that is used to build the underlying data model.
2. SQL Server Databases
Jet Analytics leverages SQL Server to store and process relational data. These databases can be hosted on-premise or in the cloud via Azure SQL Database.
3. Analysis Services
Jet Analytics leverages Analysis Services to process and store cubes. Cubes are analytical databases that are optimized for read access and rich data analysis. Analysis Services comes in several different configurations:
- SSAS Multidimensional
- Required if using on-prem active directory only
- SSAS Tabular
- Can be used if using on-prem active directory
- Can be used if using Hybrid identity model
- Azure Analysis Services
- (Cloud option)
- Required if using Azure AD cloud-only identify model
There are pros and cons to each configuration and the one you will choose will depend on a variety of factors such as cost, in-house expertise, existing infrastructure, etc..
Jet Hub (Optional Component)
The Jet Hub is an internal IIS website that is used to facilitate multiple features such as:
- Storing an easily accessible report repository
- Single location to store Jet Reports and Dashboards
- Simplifies sharing of reports with others within the organization, along with scheduling options
- Allows for simplified administration of users and data source access
Jet Reports (Optional Component)
Jet Reports is a front-end client tool. This tool is used to consume data from one or more sources. This tool is usually installed on a client’s local machine and allows users to pull live data from Cubes/DWH to Excel.
SQL Server Management Studio
Access to SQL Server Management Studio is required during the installation and is useful during development.
Option 1: On-Prem Deployment
The on-prem deployment option may be best suited for those organizations that already have local resources that they can leverage. For example, if the organization already has an instance of SQL Server or existing servers that they can re-purpose. This option may also be ideal for those organization that are not yet ready to leverage Microsoft Azure. A typical on-prem deployment would look like this:
Single Box – Deployment
In this configuration, we have a dedicated server for all the Jet components.
- Jet Data Manager
- SQL Server
- Database Engine
- SSAS Multidimensional or SSAS Tabular
- SQL Server Integration Services
- Jet Hub
- Excel / Jet Reports
- SQL Server Management Studio
Scaled out – Deployment
In this configuration the components are scaled out among multiple servers.
SQL Box
- SQL Server
- Database Engine
- SSAS Multidimensional or SSAS Tabular
- SQL Server Integration Services
- App Server
- Jet Data Manager
- Excel / Jet Reports
- SQL Server Management Studio
Option 2: Cloud Deployment
The cloud deployment option may be best suited for those organizations that may not have local resources that they can leverage. For example, if the organization does not want to invest in local hardware, licensing, etc. they may opt to leverage Microsoft Azure. A typical cloud deployment would like this:
- Azure VM to host the Jet Data Manager
- Excel / Jet Reports would also reside on Azure VM
- SQL Server Management Studio
- Azure SQL Database(s)
- Azure Analysis Services
- Azure Active Directory
Option 3: Hybrid Deployment
An organization also has the option of a hybrid deployment. In this deployment some of the components will reside on-prem, while others will reside in Azure. A typical hybrid deployment may look like this:
On-Prem components
- On-prem windows server for:
- Jet Data Manager
- Jet Hub
- Excel / Jet Reports
- SQL Server Management Studio
Azure Components
- Azure Analysis Services
- Azure SQL Databases
Comments