Oracle RAC Database Architecture Overview

Oracle Database can be configured in two basic ways: as a single instance (SI) database (non-cluster) or a Real Application Clusters (RAC) database. A single instance database has the Oracle software, database files, memory, and processes on one server. (link to database technical architecture diagram). A single-instance database has a one-to-one relationship between the instance (memory and processes) and the database (database files). Oracle RAC environments, however, have a one-to-many relationship between the database and instances. An Oracle RAC database can have up to 100 instances, all of which access one database. All the servers for a single database in the Oracle RAC environment must belong to the same Oracle Real Application Cluster.

Fig 1.1: Real Application Clusters (RAC) Overview

All application clusters have some common features, each cluster has Grid Infrastructure (GI) installed locally on each node. Grid Infrastructure includes several products: Automatic Storage Management, Oracle Clusterware, and others.

Each instance in an Oracle RAC cluster accesses the database files stored on shared storage. Each Node is also available and requires local storage that is used to store GI software and Oracle databases.

Each node needs at least a public interface and a private interface. Each network may use multiple network interface cards to increase bandwidth, availability, or fault tolerance. Oracle Corp. recommends that each network use multiple network interface cards (NICs) per node and multiple network switches in each network to avoid a single point of failure.

A cluster consists of several connected computers or servers that appear as they are a single server to end users and applications. Oracle RAC has an Oracle database option that allows you to connect to Oracle databases. Oracle RAC uses Oracle Clusterware for infrastructure to connect multiple servers together to work as one system

What is Oracle Real Application Clusters

• Multiple instances of accessing the same database
• One instance per node
• Physical or logical access to each database file
• Software-controlled data access
• Multiple instances have the same OS, the same database version, and the same network with shared storage.

You may see a few additional background processes associated with a RAC instance than you would with a single instance database. These processes are primarily used to maintain database consistency among each instance. 

They manage what is called the global resources:

    • LMON: Global Enqueue Service Monitor
    • LMD0: Global Enqueue Service Daemon
    • LMSx: Global Cache Service Processes, where x can range from 0 to j
    • LCK0: Lock process

    • DIAG: Diagnosability process

At the cluster level, you find the main processes of Oracle Clusterware. They provide a standard cluster interface on all platforms and perform high-availability operations. You find these processes on each node of the cluster

    • CRSD and RACGIMON: Are engines for high-availability operations

    • OCSSD: Provides access to node membership and group services

    • EVMD: Scans callout directory and invokes callouts in reaction to detected events

    • OPROCD: Is a process monitor for the cluster

Several tools are used to manage the various resources available on the cluster at a global level. These resources are the Automatic Storage Management (ASM) instances, the RAC databases, the services, and node applications. Some of the tools that you use throughout this post are Server Control (SRVCTL), DBCA, CRSCTL, ASMCMD.

Fig 1.2: Oracle Real Application Clusters (RAC) Architecture

RAC stands for Real Application Clusters, which is an Oracle database feature that allows multiple instances to access a single database, sharing the workload and providing high availability and scalability. 

The architecture of RAC involves several key components:

Clusterware: This is the foundation of the RAC architecture. It provides the infrastructure for node membership, communication, and cluster coordination. Oracle Clusterware ensures that all nodes in the cluster can work together seamlessly.

Cluster Interconnect: It's the private network used by cluster nodes for communication. This network is crucial for RAC as it ensures fast and reliable communication between instances, which is essential for maintaining data consistency and synchronization.

Shared Storage: All nodes in the RAC cluster must have access to shared storage where the database files reside. This shared storage can be in the form of a SAN (Storage Area Network) or NAS (Network Attached Storage),and it allows any instance in the cluster to access and modify the database files as needed.

Oracle Database: Each node in the cluster runs an instance of the Oracle Database software. These instances work together to provide access to the same database. Clients connecting to the database can be directed to any instance in the cluster, and Oracle's Cache Fusion technology ensures that data consistency is maintained across instances.

Global Cache Service (GCS): GCS is responsible for managing the global cache in a RAC environment. It coordinates access to data blocks across multiple instances, ensuring that each instance has the most up-to-date version of a data block when needed.

Oracle Services: RAC uses Oracle Services to provide high availability and workload management. Services define the characteristics of connections to the database, such as load balancing and failover behavior. By using services, administrators can control which nodes handle specific types of workloads and how connections are redirected in case of a node failure.

Automatic Storage Management (ASM): ASM is a feature provided by Oracle that simplifies storage management for database files. It abstracts the underlying storage hardware and provides features such as disk mirroring and striping, making it easier to manage storage in a RAC environment.

Overall, the RAC architecture is designed to provide high availability, scalability, and performance for Oracle databases by distributing workload across multiple nodes in a cluster and ensuring that data remains consistent and accessible even in the event of node failures.

Oracle RAC History

The first Oracle cluster database was released with Oracle 6 for the digital VAX, this was the first cluster database on the market. With Oracle 6.2 Oracle Parallel Server (OPS) was born, which used Oracle's own DLM (Distributed Lock Manager). Oracle 7 used vendor-supplied clusterware but this was complex to set up and manage, Oracle 8 introduced a general lock manager and this was a direction for Oracle to create its own clusterware product. Oracle's lock manager is integrated with Oracle code with an additional layer called OSD (Operating System Dependent), this was soon integrated within the kernel and became known as IDLM (Integrated Distributed Lock Manager) in later Oracle versions. Oracle Real Application Clusters 9i (Oracle RAC) used the same IDLM and relied on external clusterware software (Sun Cluster, Veritas Cluster, etc).