Cluster deployment

Clustering Spotfire Statistics Services can offer benefits for an expanding project or organization. When you consider deploying a cluster, review the network requirements and potential points of failure as well as the benefits.

Benefits of Clusters

Benefit Description
Expands to meet growth. You are not required to identify a single computer that will meet growth requirements. You can start conservatively and grow the system as needed.
Provides failover. Jobs running on a failed node are marked as failed. You can resubmit these jobs without needing to bring the failed node back online.
Offers high availability. You do not need to rely on a single computer. If a Spotfire Statistics Services node fails, other nodes continue processing jobs.
Offers high reliability. You can identify a growing job queue easily. Adding a Spotfire Statistics Services node is a simple process. Once the service starts, the new node immediately begins processing jobs.
Provides load balancing. Jobs are distributed among the Spotfire Statistics Services nodes to ensure optimal resource utilization.

Considerations for Clustering

Consideration Description
Requires a fast network. The cluster depends heavily on the network used by the systems in the cluster. Additional planning might be necessary to ensure the systems use appropriate interconnects and they are in close proximity.
Includes potential points of failure. The basic cluster deployment includes the following three potential points of failure.
  • Database server used for the job database.
  • File system used for the shared cluster configuration store.
  • Single point for handling incoming requests.
Spotfire Statistics Services single-manager server with workers.

You can eliminate these points of failure, but you might need additional planning and configuration. In the case of the job database and the shared cluster configuration store, most database servers and file systems offer solutions for failover and high availability. For the consideration of single point for handling incoming requests, you might consider using a replicated load balancer with two Manager nodes.

Spotfire Statistics Services two manager servers with load balancer.


Cluster Configuration Requirements

In a Spotfire Statistics Services cluster, each node must share the same configuration store and job database. It is critical that the results generated by a node are accessible to other nodes and clients in the event that node is not available. To achieve this, the configuration store, job database, and the cluster storage must be centralized.

The spserver.properties file contains the key configuration properties for Spotfire Statistics Services. See Server properties for more information.

  • Cluster scaling design
    As you plan a Spotfire Statistics Services cluster deployment, consider whether use the vertical or horizontal scaling design.
  • Manager nodes
    To implement a Spotfire Statistics Services cluster, you configure Manager nodes and Worker nodes.
  • Cluster installation checklist
    Before you establish the Spotfire Statistics Services cluster, confirm that you meet all of the requirements for a successful deployment.
  • Cluster file share
    A Spotfire Statistics Services cluster must have a file share established prior to the cluster installation.
  • Installing the first manager node on Windows
    Establish your Spotfire Statistics Services cluster by installing the first Manager node.
  • Installing the first manager node on UNIX or Linux
    Establish your Spotfire Statistics Services cluster by installing the first Manager node.
  • Running the installer
    This procedure walks you through running the Spotfire Statistics Services installer as you add the first node in a new cluster.
  • Migrating your Spotfire Statistics Services configuration and data
    Upgrading from a previous release of Spotfire Statistics Services requires a new installation. However you can retain and migrate configuration and data.
  • Post-installation configuration settings
    On both Windows and UNIX/Linux, you must make a few configuration adjustments to the Spotfire Statistics Services cluster installation and configuration, and you must configure the job database.
  • Start and validate the service
    After you have completed the Spotfire Statistics Services installation the post-installation configuration steps, you are ready to start the service.
  • Add nodes
    When you finish installing and verifying the first Spotfire Statistics Services Manager node, the settings and configuration for the cluster are complete. Now, you can add additional Manager and/or Worker nodes.
  • Load balancer
    Use a load balancer with your Spotfire Statistics Services cluster deployment to distribute job requests equally, or if you have multiple Manager nodes.
  • Automate cluster installations
    If you want to install Spotfire Statistics Services on several computers, you can perform a silent Spotfire Statistics Services installation by providing a response file to the installer. You can use this approach to automate Spotfire Statistics Services installations.