Meridian Enterprise Server Clusters

In environments that place high demands on publishing renditions, multiple Meridian Enterprise Server computers can be configured to act as one system to improve rendering performance and to simplify the configuration. Clustering provides no benefits for other system functions such as repository performance.

Note:
  • Meridian Enterprise Server clustering does not improve performance appreciably for publishing jobs that do not perform document rendering.

  • Each Meridian Enterprise Server computer in the cluster will claim an appropriate license.

You should consider Meridian Enterprise Server clustering when:

  • More than one rendering module will be used for many documents. Each of the rendering modules can be configured on a separate Meridian Enterprise Server computer dedicated to that rendering module.

  • Significant numbers of large documents will be rendered with different rendering modules. Each of the rendering modules can be configured on a separate Meridian Enterprise Server computer dedicated to that rendering module.

  • Only modest hardware resources are available for one Meridian Enterprise Server computer but multiple computers are available. Multiple Meridian Enterprise Server computers can be configured with the same rendering modules and groups of users can be assigned different publishing jobs that specify which Meridian Enterprise Server computer to process their jobs. This effectively limits the maximum load on any one Meridian Enterprise Server computer.

  • Large document renditions must be available as soon as possible after they are submitted for rendering. Each of the rendering modules can be configured on a separate Meridian Enterprise Server computer dedicated to that rendering module. Alternatively, multiple Meridian Enterprise Server computers can be configured with the same rendering modules and groups of users can be assigned different publishing jobs that specify which Meridian Enterprise Server computer to process their jobs.

For example, if publishing job A specifies the AutoCAD rendering module and job B specifies the Inventor rendering module and only one Meridian Enterprise Server computer is configured, rendering many large files can take considerable time and place a high demand on the hardware resources as AutoCAD and Autodesk Inventor are repeatedly and alternately started to perform rendering.

Meridian Enterprise Server supports clustering by allowing each Meridian Enterprise Server computer in the cluster to run specific publishing jobs. All Meridian Enterprise Server computers in the cluster share the same publishing database and Publisher Queue. One Meridian Enterprise Server computer in the cluster, called the primary, hosts the Meridian Enterprise Server web service that delegates publishing jobs to the other computers, called nodes, for rendering. The web services on the nodes are not used. The primary Meridian Enterprise Server computer can also act as a node of itself and also perform rendering.

In the preceding example, all client computer job submissions, queue management, logging, and related functions are processed by the Meridian Enterprise Server primary node. Jobs that specify the node where the AutoCAD rendering module is installed (job A in the example) run there. Jobs that specify the node where the Autodesk Inventor rendering module is installed (job B in the example) run there, and so on for additional job definitions. Each node runs its jobs asynchronously from the other computers, thereby allowing parallel rendering. Each publishing job can specify the same or a different rendering module.

Note:

Meridian does not perform load balancing, also known as pooling, between the computers in the cluster. This means that the Meridian Enterprise Server primary node cannot delegate publishing jobs to the next available node on a first-come, first-served basis. All job submissions for the same job definition will be executed on the node where the rendering module is installed regardless of how many other nodes are available. For this reason, if you want to distribute jobs roughly equally between multiple nodes, the jobs must be divided among separate nodes or divided among Meridian Enterprise Server users (instruct groups of users to select different jobs in Meridian Enterprise).

2024