How do you implement load balancing and failover in Oracle RAC?

1 Load Balancing Methods

Load balancing is the process of distributing the workload among the available servers to optimize resource utilization and response time. Oracle RAC supports two types of load balancing: server-side and client-side. Server-side load balancing is performed by the clusterware, which assigns incoming requests to the least loaded server based on various metrics. Client-side load balancing is performed by the application or the connection manager, which can use different algorithms to select the best server for each connection. You can configure both types of load balancing using parameters, services, and policies.

2 Failover Methods

Failover is the process of switching to an alternative server or instance when the primary one fails or becomes unavailable. Oracle RAC supports two types of failover: transparent and fast application notification (FAN). Transparent failover is handled by the clusterware, which automatically reconnects the client to another server without any intervention. FAN failover is handled by the application or the connection manager, which receives notifications from the clusterware about the status of the servers and instances, and can perform actions such as reconnecting, retrying, or aborting. You can configure both types of failover using parameters, services, and policies.

3 Parameters for Load Balancing and Failover

Parameters are settings that control the behavior of the database, the clusterware, and the client. You can use parameters to enable or disable load balancing and failover, specify the load balancing method and algorithm, set the failover type and timeout, and define the service quality and performance goals. Some of the parameters that affect load balancing and failover are: CLUSTER_DATABASE, REMOTE_LISTENER, LOAD_BALANCE, FAILOVER_MODE, TAF_POLICY, and SERVICE_TIME.

4 Services for Load Balancing and Failover

Services are logical entities that represent a group of applications or users that access the database. You can use services to define the workload characteristics, the availability requirements, and the performance objectives for each service. You can also assign services to specific servers or instances, and enable or disable them dynamically. Some of the attributes that affect load balancing and failover are: PREFERRED_INSTANCES, AVAILABLE_INSTANCES, CLB_GOAL, FAILOVER_TYPE, FAILOVER_METHOD, and FAILOVER_RETRIES.

5 Policies for Load Balancing and Failover

Policies are rules that govern how the clusterware manages the resources and the workload of the database. You can use policies to automate the scaling and the recovery of the servers and instances based on the demand and the events. You can also use policies to balance the load and the failover across the servers and instances according to the service quality and performance goals. Some of the policies that affect load balancing and failover are: SERVER_POOLS, INSTANCE_CAGING, INSTANCE_GROUPS, and RUN_TIME_LOAD_BALANCING.

6 Features for Load Balancing and Failover

Features are functionalities that enhance the performance, reliability, and usability of the database. You can use features to optimize the connection establishment, the data access, and the query execution across the servers and instances. You can also use features to monitor and diagnose the health and the performance of the database. Some of the features that affect load balancing and failover are: SCAN, VIP, ONS, GNS, FCF, AQ, GV$, and AWR.

7 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?