|Oracle9i Real Application Clusters Deployment and Performance
Release 2 (9.2)
Part Number A96598-01
A wait event that is signaled when a process has to wait for a block to become available because another process is obtaining a resource for this block.
A wait event that is signaled when a process cannot get a buffer because another process is using the buffer at that moment.
A diskless cache coherency mechanism in Real Application Clusters that provides copies of blocks directly from a holding instance's memory cache to a requesting instance's memory cache.
A set of instances that cooperates to perform the same task.
The generic term for an Oracle Real Application Clusters database.
See cluster database.
See cluster database.
An operating system-dependent component that discovers and tracks the membership state of each node by providing a common view of membership across the cluster. The CM also monitors process health, specifically the health of the database instance. The Global Enqueue Service Monitor (LMON), a background process that monitors the health of the Global Cache Service (GCS), registers and de-registers from the CM.
A feature that balances the number of active connections among various instances and shared server dispatchers for the same service.
A client connect request is forwarded to another listener if the first listener is not responding. Connect-time failover is enabled by service registration, because the listener knows whether an instance is up prior to attempting a connection.
Consistent gets is a statistic showing the number of buffers that are obtained in consistent read (CR) mode.
The Global Cache Service (GCS) ensures that a consistent read block (also known as the master copy data block) is maintained. The consistent read block is the master block version that records information about all changes to a block. It is held in at least one System Global Area (SGA) in the cluster if the block is to be changed. If an instance needs to read the block, then the current version of the block may reside in many buffer caches as a shared resource. Thus, the most recent copy of the block in all System Global Areas contains all changes made to that block by all instances, regardless of whether any transactions on those instances have committed.
The Oracle Enterprise Manager Console gives you a central point of control for the Oracle environment through an intuitive graphical user interface (GUI) that provides powerful and robust system management.
A file that records the physical structure of a database and contains the database name, the names and locations of associated databases and online redo log files, the timestamp of the database creation, the current log sequence number, checkpoint information and various other records about the database's structure and health.
The number of CR blocks shipped from the instance that has a block in exclusive access mode to the instance requesting a CR version of this block.
A statistic that quantifies a wait that is incurred whenever Oracle re-submits a consistent read request when Oracle detects that the holding instance is no longer available.
A statistic that shows the number of current buffers obtained in exclusive mode for DML.
A statistic that shows the number of current buffers obtained for a read.
Also known as forced writes, a statistic showing the number of writes that an instance has to perform to disk to make a previously exclusively held block available for another instance to read into its buffer cache. DBWR cross-instance writes are practically eliminated with Cache Fusion, unless you specify a value greater than 0 (zero) for the
Database and application environments that help with decision support or data warehouse systems.
A server that requires a dedicated server process for each user process. There is one server process for each client. Oracle Net sends the address of an existing server process back to the client. The client then resends its connect request to the server address provided. Contrast this with the shared server.
Specifies the number of processes, or threads, used in parallel execution. Each parallel process or thread can use one or two parallel execution processes depending on the SQL statement's complexity.
Pointers to global resources. To perform operations on global enqueue service resources, the process first needs to acquire a DFS handle.
A process that enables many clients to connect to the same server without the need for a dedicated server process for each client. A dispatcher handles and directs multiple incoming network session requests to shared server processes. See also shared server.
The number of flow-control (nullreq and nullack) messages that are sent by the LMS process.
The number of flow-control (nullreq and nullack) messages received by the LMD process.
The forced writing of a data block to disk by one instance when the data block is requested by another instance for a DML operation. Forced Writes are practically eliminated in Oracle9i with Cache Fusion, but they remain relevant if you specify 1:1 or 1:n releasable or fixed resources with the
GC_FILES_TO_LOCKS parameter. In this case, Cache Fusion is disabled for that tablespace.
A wait event that only can occur during startup or shutdown of an instance when the LMS process finalizes its operations.
A wait event that occurs whenever a session has to wait for an ongoing operation on the resource to complete.
A wait event that occurs whenever a session waits for the AST to complete for a canceled block access request. Cancelling the request is part of the Cache Fusion Write Protocol.
A statistic showing resource converts of buffer cache blocks. This statistic is incremented whenever GCS resources are converted from Null to Exclusive, Shared to Exclusive, or Null to Shared.
The accumulated time that all sessions require to perform global conversions on GCS resources.
A statistic that is incremented whenever a resource operation times out.
The time waited for a log flush when a CR request is served. Once LGWR has completed flushing the changes to a buffer that is on the log flush queue, LMS can send it. It is part of the serve time.
When a process requests a consistent read for a data block that is not in its local cache, it sends a request to another instance. Once the request is complete, in other words, the buffer has been received, Oracle increments the statistic.
A statistic that records the total time required for consistent read requests to complete. In other words, it records the accumulated round-trip time for all requests for consistent read blocks.
A statistic showing the number of requests for a consistent read block served by LMS. Oracle increments this statistic when the block is sent.
A statistic showing the time that the LMS process requires to create a consistent read block on the holding instance
A statistic showing the time required by LMS to initiate a send of a consistent read block. For each request, timing begins when the block is sent and stops when the send has completed. This statistic only measures the time it takes to initiate the send; it does not measure the time elapsed before the block arrives at the requesting instance.
Await event that occurs when a session waits for the acquisition interrupt to complete for a canceled CR request. Cancelling the CR request is part of the Cache Fusion write protocol.
A wait event that occurs whenever a process has to wait for a pending CR request to complete. The process waited for either shared access to a block to be granted before reading the block from disk into the cache, or it waited for the LMS of the holding instance to send the block.
A statistic showing the time it takes to flush the changes to a block to disk, otherwise known as a forced log flush, before the block is shipped to the requesting instance
A statistic showing the time it takes to the pin the current block before shipping it to the requesting instance. Pinning a block is necessary to disallow further changes to the block while it is prepared to be shipped to another instance.
A statistic showing the number of current blocks received from the holding instance over the interconnect.
A statistic showing the accumulated round-trip time for all requests for current blocks
A statistic showing the time it takes to send the current block to the requesting instance over the interconnect.
A statistic showing the number of current blocks shipped to the requesting instance over the interconnect
A statistic showing when a wait event that occurs if Oracle must wait after it detects that the local element free list is empty.
A statistic showing the number of times Oracle found the resource element free list empty.
A statistic showing the number of buffer gets that result in opening a new resource with the GCS.
A statistic showing the accumulated time of all sessions needed to open a GCS resource for a local buffer.
Initialization parameters that determine the size of the collection of global that protect the database buffers on all instances.
A wait event that occurs whenever a session has to wait for a resource conversion to complete.
A wait event that occurs whenever a session has to wait for this resource conversion to complete.
A wait event that occurs when a session has to wait for receiving permission for shared access to the requested resource.
A wait event that occurs when a session has to wait for receiving a exclusive access to the requested resource.
A wait event that occurs whenever a session has to wait for this resource conversion to complete.
A wait event that can occur when a process waits for an acquisition interrupt before Oracle closes a resource element.
A wait event that occurs when a session must wait for the acquisition interrupt to complete for a canceled predecessor read request. Cancelling a predecessor read request is part of the Cache Fusion write protocol.
A wait event that occurs whenever Oracle fails to prepare a buffer for a consistent read or Cache Fusion request, and when Oracle cannot ignore or skip this failure.
The process that implements Cache Fusion. It maintains block modes for blocks in the global role and is responsible for block transfers among instances. The Global Cache Service accomplishes these tasks using background processes such as the Global Cache Service process (LMS) and the Global Enqueue Service process (GES).
The processes that handle remote Global Cache Service (GCS) messages. Real Application Clusters provides for up to 10 Global Cache Service Processes. The number of LMSn varies depending on the amount of messaging traffic among nodes in the cluster. The LMSn handle the acquisition interrupt and blocking interrupt requests from a remote instance for Global Cache Service resources. For cross-instance consistent read requests, LMSn creates a consistent read version of the block and sends it to the requesting instance. LMSn also controls the flow of messages to and from remote instances.
The full name of the database that uniquely identifies it from another database. The global database name is of the form
database_name.database_domain, for example, sales.us.acme.com.
This service coordinates enqueues that are shared globally.
The resource agent process that manages Global Enqueue Service resource requests. The LMD process also handles deadlock detection Global Enqueue Service requests. Remote resource requests are requests originating from another instance.
The background process that monitors the entire cluster to manage global resources. LMON manages instance and process expirations and the associated recovery for the Global Cache and Global Enqueue Services. In particular, LMON handles the part of recovery associated with global resources. LMON-provided services are also known as cluster group services (CGS).
A statistic showing the number of resources that Oracle converted from an incompatible mode.
A statistic showing the number of GCS resources that Oracle must open synchronously. Sync gets are mostly for GES resources (for example, library cache resources).
A statistic showing the number of GES resources that Oracle must open asynchronously. Async gets are only used for GES resources and include the number of global cache gets.
A statistic showing the accumulated time for all GES resources that Oracle needed to open.
The number of GES resources that Oracle converted from an incompatible mode. Sync converts occur mostly for GES resources.
A statistic showing the accumulated time for all global lock sync converts and global lock async converts.
A database that has both OLTP and Data Warehouse processing characteristics.
A file with parameter settings that initialize the database (
.ora). In the case of Real Application Clusters, it initializes the instances within a cluster (
initsid.ora). The default single initialization parameter file is known as
The combination of the System Global Area (SGA) and each process for the Oracle database. The memory and processes of an instance manage the associated database's data and serve the database users. Each instance has unique Oracle System Identifier (SID), instance name, instance number, rollback segments, and thread ID.
Instance groups limit the number of instances that participate in a parallel operations. You can create any number of instance groups, each consisting of one or more instances. You can then specify which instance group is to be used for any or all parallel operations. Parallel execution servers will only be used on instances that are members of the specified instance group.
Represents the name of the instance and is used to uniquely identify a specific instance when several instances share common service names. The instance name is identified by the
INSTANCE_NAME parameter in the initialization parameter file. The instance name is identical to Oracle System Identifier (SID).
A number that associates extents of data blocks with particular instances. The instance number enables you to start up an instance and ensure that it uses the extents allocated to it for inserts and updates. This ensures that it does not use space allocated for other instances. The instance cannot use data blocks in another free list unless the instance is restarted with that instance number.
You can use various SQL options with the
INSTANCE_NUMBER initialization parameter to associate extents of data blocks with instances.
The instance number is depicted by the
INSTANCE_NUMBER parameter in the instance initialization file,
The communication link between the nodes.
A high-speed operating system-dependent transport component. The IPC transfers messages between instances on different nodes. Also referred to as the interconnect.
A listener configuration file that identifies the protocol addresses on which the listener is accepting connection requests and the services the listener listens for.
Load balancing is the even distribution of active database connections among instances. In the context of parallel execution, load balancing refers to the distribution of parallel execution server processes to spread work among the CPUs and memory resources.
A statistic showing the number of up-converts from Null to Shared.
A statistic showing the number of global lock sync gets and global lock async gets per transaction.
A statistic showing the number of global local sync converts and global lock async converts per transaction.
The number of messages intended to be sent directly but that are instead queued and delivered later by LMD/LMS.
The number of messages received by the LMD process.
The number of messages sent directly by Oracle processes.
The number of messages explicitly queued by Oracle processes.
A node is machine on which one or more instances reside.
Operating system context switches occur when a thread's time allotment has elapsed, when a thread with a higher priority has become ready to run, or when a running thread needs to wait, for example, for I/O to complete.
Software that consists of several software components developed by Oracle or other vendors. The OSD layer maps the key operating system/cluster-ware services required for proper operation of Real Application Clusters.
A system management tool that provides an integrated solution for centrally managing your heterogeneous environment. Oracle Enterprise Manager combines a graphical console, management server, Oracle Intelligent Agent, repository database, and tools to provide an integrated, comprehensive systems management platform for managing Oracle products.
A suite of GUI tools that make up the Oracle Enterprise Manager product.
A process that runs on each of the node that functions as the executor of jobs and events sent by the console by way of the Management Server. The Oracle Intelligent Agent ensures high availability since the agent can function regardless of the status of the Console or network connections.
A software component that enables connectivity. It includes a core communication layer called the Oracle Net foundation layer and network protocol support. Oracle Net enables services and their applications to reside on different computers and communicate as peer applications. The main function of Oracle Net is to establish network sessions and transfer data between a client machine and a server or between two servers. Once a network session is established, Oracle Net acts as a data courier for the client and the server.
An add-on application for Oracle Enterprise Manager that offers a variety of tabular and graphic performance statistics for Real Application Clusters. The statistics represent the aggregate performance for all instances.
A breakthrough architecture that enables clusters to access a shared database. Real Application Clusters includes the software component that provides the necessary Real Application Clusters scripts, initialization files, and datafiles to make the Oracle9i Enterprise Edition an Oracle9i Real Application Clusters database.
A name that identifies a specific instance of a running pre-release 8.1 Oracle database. For a Real Application Clusters database, each node within the cluster has an instance referencing the database. The database name, specified by the
DB_NAME parameter in the
.ora file, and unique thread ID make up each node's SID. The thread ID starts at 1 for the first instance in the cluster, and is incremented by 1 for the next instance, and so on.
Oracle9i Enterprise Edition is an object-relational database management system (ORDBMS). It provides the applications and files to manage a database. All other Real Application Clusters components are layered on top of the Oracle9i Enterprise Edition.
A feature that automatically controls values for all parameters related to parallel execution. These parameters affect several aspects of server processing, namely, the degree of parallelism (DOP), the adaptive multiuser feature, and memory sizing. Initialize and automatically tune parallel execution by setting the initialization parameter
Multiple processes operating together to complete a single database transaction. Parallel execution works on both single instance and cluster database Oracle installations. Parallel execution is also referred to parallel query.
A statistic showing the number of disk reads that had to be performed when a request for a data block could not be satisfied from a local cache.
The number of write I/Os performed by the DBWNn processes. This number includes the number of DBWR cross instance writes (forced writes) in Oracle9i when
GC_FILES_TO_LOCKS is set. Setting
GC_FILES_TO_LOCKS for a particular datafile will enable the use of the old ping protocol, and will not leverage the Cache Fusion architecture.
Disks or partitions on disk drives that do not have a file system set up on them. Raw devices are used for Real Application Clusters since they enable the sharing of disks.
See raw devices.
An Oracle tool that enables you to back up, copy, restore, and recover datafiles, control files, and archived redo logs. It is included with the Oracle server and does not require separate installation. You can invoke RMAN as a command line utility from the operating system (O/S) prompt or use the GUI-based Enterprise Manager Backup Manager.
A statistic showing the number of undo blocks written to disk by DBWn as part of a forced write.
A statistic showing the number of rollback segment header blocks written to disk by DBWn as part of a forced write.
A database, such as that used by Oracle Enterprise Manager, that is a set of tables in an Oracle database, to store data to manage Real Application Clusters environments. This database is separate from any shared Real Application Clusters database on the nodes.
A preconfigured, ready-to-use database that requires minimal user input to create.
A comprehensive, integrated system management solution for managing Real Application Clusters environments. Server Management enables you to manage cluster databases in heterogeneous environments. Server Management is part of the open client/server architecture of Oracle Enterprise Manager. In addition to managing cluster databases, Server Management enables you to schedule jobs, perform event management, monitor performance, and obtain statistics to monitor Real Application Clusters performance.
A logical representation of a database, which is the way a database is presented to clients. A database can be presented as multiple services and a service can be implemented as multiple database instances. The service name is a string that is the global database name, a name comprised of the database name (
DB_NAME) and domain name (
DB_DOMAIN), entered during installation or database creation.
A feature by which the PMON process (or shared server Dispatcher processes when using shared server) automatically registers information with a listener. Because this information is registered with the listener, you do not need to configure the
listener.ora file with this static information.
A server configured to enable many user processes to share very few server processes. This means increases the number of users that can be supported. With shared server, many user processes connect to a dispatcher. Contrast this with dedicated server.
Query-centric schemas that when represented in a diagram have a fact table at the center. The fact table usually contains the data element that is central to queries operating against the schema. A fact table is often quite large and is surrounded by several dimension tables that contain data that are attributes of the data in the fact table. Star schemas simplify query development because it is intuitive as to how to join attributes in the dimension tables with the fact table data. Star schemas are best suited for data warehouse environments and are thus less useful for OTLP environments.
Refers to the interleaving of a related block of data across disks. If you properly implement striping, then it reduces I/O and improves performance. Because striping software is operating system-dependent, rely on your vendor documentation to ensure proper installation and configuration.
A set of protocols that enable cooperating computers to share resources across a network.
The number of a redo thread for an instance. Any available redo thread number can be used, but an instance cannot use the same thread number as another instance.
A class of software products that provide a transaction execution layer on top of the operating system. Transaction monitors combine database updates and submit them to a database. In doing this, the transaction monitor manages some of the consistency and correctness of the database. The monitor ensures that the rules of transaction atomicity are adhered to; updates take place completely or not at all. The advantages of using transaction monitors include increased throughput.
A runtime failover mechanism for high-availability environments, such as Real Application Clusters and Oracle Real Application Clusters Guard, that refers to the failover and re-establishment of application-to-service connections. It enables client applications to automatically reconnect to the database if the connection fails, and optionally resume a
SELECT statement that was in progress. This reconnect happens automatically from within the Oracle Call Interface (OCI) library.
A protocol similar to TCP/IP. However, it is simpler to administer. It is considered less reliable than TCP/IP because, for example, it does not guarantee message ordering.
An IPC-based (Inter-process Communication) protocol that directly accesses network hardware. As opposed to kernel-mode IPC, with user-mode IPC the protocol avoids the overhead of copying data into kernel space, making system calls, and incurring context switches.
An implementation of user mode IPC.