Planning a Load-Balancing Cluster

Overview

To plan for a new load-balancing cluster, you choose hosts, port numbers, a load-balancing algorithm, and logical server login credentials (for OLAP clusters).

Select Hosts and Port Numbers

To select the hosts that will make up your load-balancing cluster, follow these steps:
  1. Determine the number of hosts that you will need in your cluster based on your average number of concurrent users and your maximum number of concurrent users.
  2. Decide on the logical server that will contain your cluster of servers. You can use an existing logical server or create a new one using the SAS Deployment Wizard. In either case, you will have one configured server available under that logical server, as shown in SAS Management Console. Note that the additional processing load on the parent peer is not significant, so the parent peer does not have to be the highest performing host in your cluster.
  3. Select and configure the hosts that will comprise your cluster.
  4. Select and write down unique server names and unique port numbers for each host. (Unique ports are required when configuring load balancing on a single machine.) You will enter the server names when you copy and update the server configuration files between hosts. In the server names, you might want to identify the servers as load balancing and include the port number for quick reference.

Choose a Load-Balancing Algorithm

Select a load-balancing algorithm based on the choices that are available for your server type. Clusters of SAS OLAP Servers are required to use either the Least Recently Used or Grid algorithm. The Cost algorithm is recommended for clusters of SAS Workspace Servers and SAS Stored Process Servers.
The following table depicts the algorithms that are available for each server type.
Available Load-Balancing Algorithms
SAS Server Cluster
Load Balancing Algorithms
Cost
Response Time
Grid
Most Recently Used
Least Recently Used
Workspace Server
X
X
Stored Process Server
X
X
X
Pooled Workspace Server
X
X
X
OLAP Server
X
X
For more information about the load-balancing algorithms, see Understanding the Load-Balancing Algorithms.
Copyright © SAS Institute Inc. All rights reserved.