Cluster Design Process
The rest of this chapter looks at how you would design a JasperReports Server cluster to fit your needs. It assumes that your cluster will follow the traditional pattern with a load balancer and some number of identical nodes, as shown in the sample architecture. Other architectures, such as dedicated OLAP nodes or a geographically distributed cluster, are possible but beyond the scope of this chapter.
As with any software project, careful design and planning will help you meet your goals. A simplified process for designing a cluster might include the following steps:
|
1.
|
Gather cluster requirements in the following areas: |
|
•
|
Performance – Usually defined as average response time for a given load. |
|
•
|
High availability – Usually measured as percentage up-time. |
|
•
|
Scalability – The ease of adding nodes to improve performance and availability over time. |
|
2.
|
Estimate the size of your cluster to meet your requirements within your limitations such as time and budget. Sizing determines the architecture of your cluster: |
|
•
|
Load-balancing hardware |
|
•
|
Size and number of cluster nodes |
|
•
|
Shared resources, especially databases |
|
•
|
JasperReports Server configuration |
|
•
|
Hardware purchases and installation |
|
•
|
Software configuration, including JasperReports Server configuration |
|
•
|
Testing of all components individually and in the cluster architecture |
|
•
|
Administration, maintenance, and scaling procedures |
Deployment and implementation are beyond the scope of this chapter. The following sections give more details about gathering cluster requirements and sizing.
Recommended Comments
There are no comments to display.