I. Adapting Hardware
| Hardware | Adaptation | |
|---|---|---|
| Operating System | The designer supports installation on Windows and Mac OS systems The FineReport project supports deployment on Windows, Mac OS, Linux, and Unix systems. | |
| Database | Support some mainstream relational databases and multidimensional databases, such as: Oracle, SQL Server, MySQL, Access, Sybase, DB2, Postgre, Derby, HIVE | |
| Server | Support Tomcat7 and above, JBOSS6.0.0 and above, WebLogic12c and above, WebSphere8.0 and above, Resin4.0.4 and above, TongWeb6 and above, etc. Web application servers | |
| Browser | Basic Function | Recommend Chrome, FireFox and other mainstream browsers Support IE9 and above browsers IE browser is not recommended |
| New Data Entry | It is recommended to use mainstream browsers such as Chrome and FireFox IE10 and below browsers are not supported IE browser is not recommended | |
| Full Screen | It is recommended to use mainstream browsers such as Chrome and FireFox IE10 and below browsers are not supported IE browser is not recommended | |
| Extended Chart | Only supports Chrome browser | |
Note: Domestic dual-core browsers such as 360 browser, etc., are all Chrome and IE kernels. It is recommended to use the Chrome kernel in extreme speed mode.
II. FineReport Configuration Test Report
2.1 Test Target
Compare and analyze the performance under different scenarios.
Through multi-dimensional test analysis and summary, the recommended configuration under different concurrency is finally given.
2.2 Test Content
This test uses 7 high-frequency user scenarios of the product research, selects the template of the customer scenario to modify, and designs the concurrency according to the business ratio.
Under the specified response time, adjust the hardware configuration, test the recommended configuration, the minimum configuration, and the maximum number of concurrent users under the recommended configuration.
2.3 Conclusion
Note 1: The following test conclusions are based on this standard template test, and the data is for reference only. The template situation of the actual environment will be more complicated, and the user should adjust it according to his actual situation.
Note 2: Network and database SQL processing will affect the experience time. This test is only for FineReport processing time. Intranet, non-complex SQL, network speed and database factors can be ignored.
2.3.1 Test Conclusion
1) In a stand-alone scenario, when the physical memory is 64g and the JVM memory is set to 8g, 16g, and 24g, the JVM memory increases and the response time of the template decreases under the same concurrency number.
2) As the number of nodes in the cluster increases, the number of concurrency that the system can handle also increases.
3) When the concurrency of a two-node cluster is twice that of a single machine and the concurrency of a three-node cluster is three times that of a single machine, the average response time of the template is similar.
4) For single-machine and dual-node cluster scenarios, file server sharing and automatic synchronization between nodes have similar performance.
5) The platform authentication preview mode has worse performance than the mode without platform authentication preview, because different users cannot share the same result cache.
2.3.2 Recommended Configuration
| Number of registered users | Number of online users | Number of concurrent users | Recommended configuration | Minimum configuration |
|---|---|---|---|---|
| 2000-5000 | 400-1000 | 0-200 | Stand-alone CPU: 8 cores 16 threads 2GHZ JVM memory: 8GB Physical memory: 64G | Stand-alone CPU: 8 cores 16 threads 2GHZ JVM memory: 8GB Physical memory: 64G |
| 4000-10000 | 800-1500 | 200-300 | Stand-alone CPU: 8 cores 16 threads 2GHZ JVM memory: 16GB Physical memory: 64G | Stand-alone CPU: 8 cores 16 threads 2GHZ JVM memory: 8GB Physical memory: 64G |
| 6000-12000 | 1200-2500 | 300-500 | Stand-alone CPU: 8 cores 16 threads 2GHZ JVM memory: 24GB Physical memory: 64G | Stand-alone CPU: 8 cores 16 threads 2GHZ JVM memory: 16GB Physical memory: 64G |
| 8000-20000 | 1600-4000 | 500-800 | Dual node CPU: 8 cores 16 threads 2GHZ JVM memory: 16GB Physical memory: 64G | Dual node CPU: 8 cores 16 threads 2GHZ JVM memory: 16GB Physical memory: 64G |
| 10000-25000 | 2000-5000 | 800-1000 | Three nodes CPU: 8 cores 16 threads 2GHZ JVM memory: 16GB Physical memory: 64G | Dual node CPU: 8 cores 16 threads 2GHZ JVM memory: 24GB Physical memory: 64G |
| 12000-30000 | 2400-6000 | 1000-1200 | Three nodes CPU: 8 cores 16 threads 2GHZ JVM memory: 24GB Physical memory: 64G | Three nodes CPU: 8 cores 16 threads 2GHZ JVM memory: 16GB Physical memory: 64G |
Description:
1) Number of registered users: users registered in the system
2) Number of online users: the number of users online at the same time in a certain period of time (between 5%-20% of the general registered number, whichever is 20%)
3) Number of concurrent users: the number of users who send requests to the server at the same time (usually about 10%-25% of the online population)
4) Recommended configuration: the average response time of 70% of the scenes is less than 4s
5) At the lowest configuration: the average response time of 70% of the scenes is less than 8s
6) JVM memory: the memory allocated to the FineReport container Tomcat (can be set to about 80% of the physical memory)
