Scalability and Performance
Application Performance
The application’s performance is very dependent on external factors, such as ArcGIS Server configuration, hardware, software, networking, and bandwidth. The application stack is made up of a set of RESTful services and a web application. The application services do not typically query data themselves – but instead the web application will leverage an ArcGIS Server map service to query data. See Application Architecture for more details.
Since neither the web application (simple HTML site) nor the .NET MVC application (REST services) query data themselves, the biggest point of potential performance issues lies within the ArcGIS Server map service. ArcGIS Server performance is essential for the web application to perform adequately. See ArcGIS Server performance documentation for details.
Scalability
In order to accommodate a very large number of concurrent users, the application server (REST services and web app) can be replicated/scaled to be placed in a load-balanced environment. This “replicating” of services will allow for spreading the load over multiple resources. There are however a number of configuration considerations that should be taken when using "replicating" of services. You can find these configuration considerations in Configuration for Load Balanced Environments located in the Configuring the Application section.
Example Metrics
The following is a list of metrics gathered, and the hardware and software they were performed on.
Configuration 1
Description:
- Application Server (Server #1)
- Windows Server 2012
- 64-bit
- Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz, 2499 Mhz, 2 Core(s), 4 Logical Processor(s)
- 30.5 GB Physical Memory
- Installed Software: ArcGIS Server, ArcGIS Desktop, SQL Server 2012
- Configuration Database (SQL Compact Edition)
- ArcGIS Server (Server #2)
- 10.2.1
- Default configuration
- Database (Server #3)
- Oracle
Infrastructure:
Performance Metrics:
| Total Miles | Layers Queried | Features Queried | Symbology Complexity (1-10) | Total Query & Render Time (seconds) |
| 3.5 | 15 | 26 | 7.5 | 5.98 |
| 25.07 | 8 | 319 | 7.5 | 9.91 |
| 25.07 | 9 | 416 | 9 | 22.87 |
| 20 | 9 | 258 | 9 | 13.05 |
| 5 | 15 | 125 | 7.5 | 3.68 |
| 5 | 4 | 19 | 4 | 1.1 |
25 Concurrent Users Metrics
Configuration 2
Description:
- Application Server (Server #1)
- Windows Server 2012
- 64-bit
- Intel(R) Xeon(R) CPU E5-2670 v2 @ 2.50GHz, 2499 Mhz, 2 Core(s), 4 Logical Processor(s)
- 30.5 GB Physical Memory
- Installed Software: ArcGIS Server, ArcGIS Desktop, SQL Server 2012
- Configuration Database (SQL Compact Edition)
- ArcGIS Server (Server #2)
- 10.2.1
- Default configuration
- Database (Server #1)
- SQL Server 2012
Infrastructure:
Performance Metrics:
| Total Miles | Layers Queried | Features Queried | Symbology Complexity (1-10) | Total Query & Render Time (seconds) |
| 12.525 | 13 | 40 | 5 | 1.96 |
| 3 | 13 | 103 | 8.5 | 4.39 |
| 5 | 13 | 148 | 148 9 | 3.85 |
| 18.68 | 13 | 14 | 4 | 2.07 |
| 40 | 4 | 12 | 2 | 2.09 |
| 4.662 | 13 | 15 | 3 | 1.44 |
25 Concurrent Users Metrics:
