Rack unit; a standard unit of space (44.45 mm/1.750 in.) into which a server can be mounted in a rack.
A list of users, groups, or enterprise roles who have permission to access or interact with a particular content item.
The degree to which software can be used comfortably by a variety of people, including those users who require assistive technologies or those who use the keyboard instead of a pointing device.
A number that is used by system software to identify a storage location. In networking, a unique code that identifies a node to the network.
A means for mapping Internet addresses into physical MAC addresses or domain addresses.
A protocol used to associate an IP address with a network hardware address (MAC address). See also machine access code (MAC) address.
An open standard that provides operating systems with the ability to perform an orderly shutdown of the operating system when power is removed suddenly. It also provides other features such as the ability to perform discovery and configuration of hardware components.
A detailed error reporting mechanism for PCIe. See also Peripheral Component Interconnect (PCI).
A characteristic of a signal that is used to initiate an action. See also unasserted.
A log that is accessible using Oracle ILOM. This log tracks all interface-related user actions, such as user logins, user logouts, configuration changes, and password changes. This log is helpful for auditing user activity to ensure that no privilege violations have occurred.
A process that verifies the identity of a user in a communication session, or a device or other entity in a computer system, before that user, device, or other entity can access system resources.
A process that grants specific access privileges to a user, such as using a service, or which objects the user is allowed to access, and the type of access that is allowed for each object. Authorization is based on authentication and access control. See also authentication.
An Oracle diagnostic software service provided to customers who have Oracle Premium Support warranty, that automatically opens a service request, downloads telemetry data, and orders parts (if necessary) when a hardware fault is detected. Oracle diagnostic capabilities analyze the data for known issues and send corrective action to the customer.
A RAS feature that initiates a system reboot sequence that bypasses failed system components or a software failure. See also reliability, availability, serviceability (RAS).
A measurement of the percentage of time that a system is accessible by users and is providing service. See also reliability and serviceability.
A measure of the volume of information that can be transmitted over a communication link. Often used to describe the number of bits per second a network can deliver. See also bits per second (bps).
A computer without an operating system, which allows microprocessors to boot directly into single-purpose software without loading an operating system.
A device used to manage chassis environmental, configuration, and service functions, and receive event data from other parts of the system. Typical functions of the BMC are to measure processor temperature, power supply values, and cooling fan status. The BMC can take autonomous action to preserve system integrity.
System software that controls the loading of the operating system and testing of hardware at system power-on. BIOS is stored in read-only memory (ROM). Also referred to as Legacy BIOS. See also Unified Extensible Firmware Interface (UEFI).
A process that automatically runs at system power-on to control the first stage of system initialization and hardware tests. Boot loader software then transfers control to a more complex program that loads the operating system.
A system that provides client systems on the network with the programs and information that the clients need to start. See also client.
In the client-server communications model, the client is a process that remotely accesses resources of a compute server, such as compute power or large memory capacity. At times, a server itself can be a client of some other server or service. See also client-server model.
A network architecture that consists of a database server that uses a relational database management system (DBMS) to quickly respond to user queries (directed from the client). See also client.
A restart of the server from a completely powered-off state. See also warm reset.
A description of components that must be serviced only when the server is completely powered off and AC power cords are disconnected from the server.
A user interface that enables you to type executable instructions at a user prompt.
A logical device or physical device that displays system messages and is used for interactive purposes. A physical console might be a local alphanumeric (ASCII) terminal or local graphics monitor. A logical device, also known as a system console, gives users privileged access to the operating system to administer hardware, services, operating systems, storage, and more.
A network component that combines Fibre Channel HBA and Ethernet NIC functionality on the same adapter card with one or more physical Ethernet ports. Used by FCoE technology. See also Fibre Channel over Ethernet (FCoE).
The international standard for time. UTC was formerly called Greenwich Meridian Time (GMT). UTC is used by NTP servers to synchronize systems and devices on a network. See also Network Time Protocol (NTP).
A file created by Oracle Solaris or a Linux operating system when a program malfunctions and terminates. The core file holds a snapshot of memory, taken at the time the fault occurred. This file can be used to determine the cause of the malfunction.
An error that can be corrected by error correction code (ECC). See also error correction code (ECC) and uncorrectable error (UE).
In certain servers (such as the Oracle Servers X7-8 and X8-8), the modules that contains the processor(s) (CPUs) and system memory.
A system component that the user can replace without special training and typically without special tools. See also field-replaceable unit (FRU).
Equipment that acts like a terminal; most often an actual terminal or terminal emulation software running on a laptop computer.
A computer that runs client applications and databases, and is connected to other systems over a network. See also client-server model.
The name that the system uses to identify a device. For example, /dev/rst0 (or only rst0) might be the device name for a tape drive.
A certification of the source of digital data that can ensure data integrity and detect data modificaitons. The signature is a number derived from a public key cryptographic process.
The transfer of data directly into memory without supervision of the processor. The data is passed on the bus directly between the memory and another device. See also direct virtual memory access (DVMA).
A mechanism to enable a device on the SBus to initiate data transfers between it and other SBus devices, such as system memory. See also direct memory access (DMA).
A backplane designed to hold multiple disks, such as 8-disk or 12-disk implementations.
A client on a network that relies on a server for resources, such as files, but has its own local disk storage. Some of a diskfull client's files are local and other files are remote. See also diskless client.
A client on a network that relies on a server for all of its disk storage. See also diskfull client.
The name that is assigned to a group of systems on a local network that share administrative files. A domain name is required for the Network Information Service (NIS) database to work properly.
A network service that translates domain names into IP addresses. This is a specific network server that maintains the list of all host names in a domain.
Describes a class of memory module (DIMM) that achieves twice the bandwidth of single rate modules by transferring data on both the rising and falling edges of the clock signal without increasing the clock frequency. See also quad data rate (QDR).
A small circuit board (card) that holds memory chips (64-bit path).
A protocol that enables a DHCP server to assign IP addresses dynamically to systems on a Transmission Control Protocol/Internet Protocol (TCP/IP) network.
A sudden, potentially damaging flow of electricity that occurs when two differently charged objects come into contact, for example when a human operator without antistatic protection touches a component.
A type of code that adds extra bits to words or double words, corrects all single-bit errors, and detects all double-bit errors.
An industry-standard type of local area network (LAN) that enables real-time communication between systems connected directly through cables.
The most widely used LAN access method defined by IEEE 802.3 standard; uses standard RJ-45 connectors and telephone wire. 100BASE-T is referred to as Fast Ethernet; 1000BASE-T is referred to as Gigabit Ethernet. See also Gigabit Ethernet.
A log, accessible using Oracle ILOM, that tracks informational, warning, or error messages about a device, such as the addition or removal of a component or the failure of a component. The properties of the events recorded in the log can include: the severity of the event, the event provider (class), and the date and time the event was logged. This log is helpful when troubleshooting the system or monitoring system performance. See also syslog and system event log (SEL).
A signal that sends a "soft" reset to the processor in a domain. XIR does not reboot the domain. An XIR is generally used to escape from a hung system so that a user can reach the console prompt.
The automatic transfer of a service from one system, or more often a subsystem, to another system to provide redundant service.
Software that can be used to manage server faults from the host operating system in a similar way that you manage faults in Oracle ILOM.
A high-speed network technology that runs primarily on optical fiber cables. See also Fibre Channel over Ethernet (FCoE).
A method to map FC frames over selected full-duplex IEEE 802.3 networks, allowing FC to leverage 10 GigabitEthernet networks while preserving the FC protocol. See also Fibre Channel (FC).
A system component that is replaceable at the customer site only by qualified Oracle Services personnel. See also customer-replaceable unit (CRU).
Circuity that includes critical system logic for the motherboard and disk backplanes.
Software that is typically used to help with the initial booting of a system and with system management.
A special type of non-volatile solid-state memory that can be electrically erased and reprogrammed, making it more economical than traditional memory modules (DIMMs).
A unique Internet name of a system, such as www.oracle.com. The FQDM includes the host server name (www), its second-level domain (oracle) and its top-level domain (com). An FQDN can be mapped to the IP address of a system.
Inputs and outputs with no predefined purpose, often used for custom system designs.
A data storage device that uses rapidly rotating, rigid magnetic disks to store and retrieve digital data in a random access manner. See also solid state drive (SSD).
A capability of a system to mask many individual points of failure or to significantly compensate for them. This type of system might be built with limited hardware and/or software components to minimize the impact of failures. Generally, this type of system is less costly than a fault tolerant system. See also failover.
A system, such as a backend server, with an assigned IP address and host name. The host is accessed by other remote systems on the network. See also local host.
A controller board that connect the I/O expansion bus to the Small Computer System Interface (SCSI) subsystem.
A supercapacitor assembly on the HBA card that provides short term backup of primary power. See also host bus adapter (HBA).
A switched fabric communications link primarily used in high-performance computing. See also InfiniBand host channel adapter (IB-HCA).
The name of a particular machine within a domain. Host names always map to a specific IP address.
The ability to install or replace a component, while the system is powered on, without interrupting the system operation. In some cases the user must prepare the system for the hot-plug operation by performing certain administrative tasks. After a new component is inserted, the user might need to instruct the system to reconfigure the device into the system. See also hot swap.
A description of components that can be serviced while the server is powered on and running in Main power mode. See also Main power mode.
Drive(s) within a storage system that are held in reserve to replace any other failed drive.
The ability to install or replace a component while the system is powered on and in operation. These components can be removed or inserted without preparing the operating system. The system either automatically recognizes the component change and configures it or requires user interaction to configure the system. In either case a reboot is not required. All hotswappable components are hot pluggable, but not all hotpluggable components are hot swappable. See also hot plug.
A command-line utility accessed by using the Oracle ILOM Diag shell that is used to check the status of system components.
Server management capability that is enabled only when the operating system is initialized and the server is functioning properly. See also out-of-band (OOB) system management.
Networking standard used in high-performance computing that provides very high throughput and low latency.
A switched fabric communications link primarily used in high-performance computing. See also host channel adapter (HCA).
A hardware-level interface specification for out-of-band management of servers over a number of different physical interconnects. Monitoring and control of hardware is implemented independent of the main processor (CPU), BIOS, and operating system. IPMI functionality includes FRU inventory reporting, system monitoring, logging, system recovery (including local and remote system resets and power-on and power-off capabilities), and alerting. See also IPMItool.
A unique 32-bit value that identifies network hosts using Transmission Control Protocol/Internet Protocol. No two network hosts can be assigned the same IP address.
A utility used to manage IPMI functions on either a local or remote system. Functions include managing FRU information, local area network configuration, sensor readings, and remote system power control.
A software development environment for writing applets and applications in the Java programming language.
An Oracle ILOM console that enables users to redirect devices (keyboard, mouse, video display, storage devices) from a local desktop to a remote host server.
A subset of the JDK that enables developers to redistribute the runtime environment. The Java runtime environment consists of the Java virtual machine (JVM), Java core classes, and supporting files. See also Java Development Kit (JDK).
The core of the operating system that manages the hardware and provides fundamental services, such as filing and resource allocation, that the hardware does not provide.
A series of interfaces that enables a system to respond to keyboard, video, mouse, and storage events.
Information stored in FRU components that is used for service entitlement and warranty coverage. It includes information such as product name, product part number, and serial number.
Chip circuits embedded on the system motherboard that can provide network connections, thereby eliminating the need for a separate NIC to access a local area network, such as Ethernet. See also network interface card (NIC).
Refers to the size of the drive bays. See also small form factor (SFF).
Technology that enables out-of-band communication with the server even when the operating system is not running. This allows users to switch the server on and off; view system temperatures, fan speeds, and so forth; and restart the system from a remote location.
A standard, extensible directory access protocol that provides a common language that LDAP clients and servers use to communicate.
A load-reduction dual inline memory module that supports higher densities than registered dual inline memory modules (RDIMMs) and contains a memory buffer (MB) chip, as opposed to a register, in order to reduce and minimize the load on the server memory bus. See also Registered dual inline memory module (RDIMM).
A computer on which a software application is running. See also host.
A device connected to a host, identified by a LUN. See also logical unit number (LUN).
A section of a formatted disk allocated by the software. Also called a partition. See also partition.
A group of physical drives configured with a RAID level. Each logical drive can be configured for a different RAID level. See also redundant array of independent disks (RAID).
A three-bit identifier used on a SCSI bus to distinguish between up to eight devices (logical units) with the same SCSI ID. See also logical device (LDEV).
Worldwide unique, 48-bit, hardware-encoded address number that uniquely identifies a piece of network hardware. See also network interface card (NIC).
The server power mode that is entered after the service processor boots into Standby power mode. In Main power mode, power is supplied to all server components, the server boots, and the operating system functions. See also Standby power mode.
A system event that impairs service, but not critically. See also critical event and minor event.
A collection of objects in a virtual database used to manage entities for Simple Network Management Protocol (SNMP).
A unit of measurement, often used when referring to memory modules (DIMMs).
A system event that does not impair current service, but which requires attention before it becomes more severe. See also major event and critical event.
The process of duplicating data from a primary location to a secondary location, so that data is still available if the primary location fails. Referred to as RAID-1 in disk arrarys. See also redundant array of independent disks (RAID) and striping.
The process of accessing a directory from a disk attached to a machine making the mount request or from a remote disk on a network. See also unmount.
A fault-tolerance and performance enhancement technique where there is more than one physical path between the processor in a computer system and its mass storage devices through the buses, controllers, switches, and bridge devices connecting them.
A higher availability option that provides two independent paths to storage and/or networks. An intermediate software layer is generally required to mask the failure of one path from the application. When both paths are functional, higher bandwidth and throughput is possible as a secondary benefit beyond higher availability. See also high availability (HA).
A network device container name. Physical device address used for remote access, configuration, and management.
A card that connects a workstation or server to a network. Also known as a network interface controller.
An Ethernet port for access to the server SP. See also service processor (SP).
A protocol that synchronizes the clocks between computer systems over packet-switched, variable-latency data networks. Often used to synchronize servers with Coordinated Universal Time (UTC). See also Coordinated Universal Time (UTC).
An addressable point or device on a network. A node can connect a computing system, a terminal, or various peripheral devices to the network.
An Oracle ILOM feature that collects power telemetry from the power supplies and communicates with the processors to affect changes in core voltage and frequency and with the memory controller to affect changes in memory utilization. Node Manager manages the processor and memory performance levels to achieve a user-set power limit that the system must not exceed. See also Oracle Integrated Lights Out Manager (ILOM).
A specification for accessing solid state drives (SSDs) that are attached through the PCIe Express bus. See also solid state drive (SSD) and PCI Express (PCIe).
A type of memory that ensures that data is not lost when system power is off.
A user with limited privileges to a managed host system. See also administrator.
An Oracle software solution that provides CLI tools to manage or monitor a system through the host operating system command line, as well as an SNMP monitoring agent.
An Oracle ILOM command shell that is used to diagnose, view, and manage fault activity on a server or other devices.
A graphical remote console feature of Oracle ILOM that enables users to redirect devices (keyboard, mouse, video display, storage media) from a desktop to a remote host server.
Oracle's implementation of lights out management (LOM). An Oracle hardware and firmware solution that enables out-of-band communication with the server even when the operating system is not running. This allows users to switch the server on and off; view system temperatures, fan speeds, and so forth; and restart the system from a remote location.
An Oracle technology for deploying operating systems and application software in a virtualization environment.
A feature of Oracle VM that allows users to manage Oracle VM Servers, virtual machines, and resources. See also Oracle VM.
A managed virtualization environment that provides a lightweight, secure, server platform for running virtual machines (Linux, Oracle Solaris, or Microsoft Windows), also known as domains. See also Oracle VM.
Server management capability that is enabled when the operating system network drivers or the server is not functioning properly. See also in-band system management.
Sections on a LDEV. Each partition can either have some space allocated to it, or can be empty. See also logical device (LDEV).
An industry-standard bus architecture that supports high-bandwidth peripherals and I/O devices.
An industry-standard bus architecture. See also PCI Express (PCIe).
A set of privileges granted or denied to a user or group that specify read, write, or execution access to a file or directory.
An hardware address that matches a memory location; programs that refer to virtual addresses are subsequently mapped to physical addresses.
A configured alert triggered by a hardware or firmware (BIOS) event. A PET is an IPMI-specific, SNMP trap, which operates independently of the operating system.
A value that shows either the input power consumed by the system or the output power provided by the PSUs.
A board that brings power from the PSUs to the motherboard. See also power distribution unit (PDU).
Two matching power units in a rack, with each unit sized for the maximum expected power load for all equipment in the rack.
A type of host firmware that runs at system startup and that tests core system components, including processors, memory, and motherboard I/O bridge integrated circuits.
An industry-standard client-server interface that enables a server to boot an operating system over Dynamic Host Protocol/Internet Protocol (TCP/IP) network using DHCP. See also Dynamic Host Configuration Protocol (DHCP).
Non-volatile memory that, once programmed, cannot be programmed again.
A mechanism whereby one system acts on behalf of another system when responding to protocol requests.
A class of memory module that can transfer up to four words of data in each clock cycle. QDR uses two clocks, one for read data and one for write data. See also double data rate (DDR).
The modular unit of measurement on which panel heights are based. A 1U system is one rack unit, which equals 44.45mm (1.750 inches). A 2U system is two rack units, which equals 88.9mm (3.5 inches).
A technology that allows remote identification of a product by the insertion of an electronic tag that can be read by radio waves.
A disturbance in the radio frequency spectrum that is generated by an external source and that effects electrical circuits.
A battery-backed component that maintains the time and date for a system, even when the system is powered off.
An operating system-level operation that performs a system shutdown followed by a system boot. Power is a prerequisite.
A set of disk drives that appear to be a single logical disk drive to an application, such as a database or file system. Different RAID levels provide different capacity, performance, availability, and cost characteristics. See also mirroring and striping.
A dual inline memory module (DIMM) with improved reliability. The RDIMM, which became available for DDR3 memory, uses a hardware register that buffers the control signals (not the application data) to the modules.
A capability that enables a system to operate continuously without failures and to maintain data integrity. Reliability influences MTBF. See also availability and serviceability.
A description of three aspects of system design that contributes to continuous operation and minimizes system downtime for services. Together RAS features provide for near continuous system operation. See also reliability, availability, and serviceability.
A feature that allows redirection of the server keyboard, video output, mouse, and storage devices through a networked system.
A European Union (EU) directive that bans the sale of certain electronic equipment containing these materials: lead, mercury, hexavalent chromium, cadmium, and certain brominated flame retardants.
In UNIX operating systems, the name of the superuser who has permissions to access any file and to carry out operations not granted to other users. Root is roughly equivalent to the administrator user name in Windows Server operating systems.
A UNIX shell program and network protocol that enables secure and encrypted log in and execution of commands on a remote system over an insecure network.
A protocol that enables client-to-server communication on a network to be encrypted for privacy. All data exchanged is encrypted with a cipher and hashed to protect it from eavesdropping and alteration. Hypertext Transfer Protocol Secure (HTTPS) uses SSL.
A computer bus primarily designed for transfer of data between a computer and storage devices (hard disk drives or optical drives).
A computer bus technology primarily designed for high transfer rates of data to and from devices, such as hard drives. SAS is faster and more versatile than SATA.
A serial port to the server SP. See also service processor (SP).
A device used by Oracle ILOM that manages chassis environmental, configuration, and service functions, and receives event data from other parts of the system. Typical functions of the SP are to measure processor temperature, power supply values, and cooling fan status. The SP can take autonomous action to preserve system integrity. See also Oracle Integrated Lights Out Manager (ILOM).
A measurement of the time required to restore a system to operation when a failure has occurred. Serviceability influences MTTR. See also reliability and availability.
A connection method to Oracle ILOM over the host data network. See also in-band system management and out-of-band (OOB) system management.
A protocol used to exchange data about network activity such as monitoring and notification of network errors and other events.
An extension to the PCIe specification that allows a PCIe device, such as a network adapter, to appear to be multiple, separate physical PCIe devices.
A form of authentication in which a user enters credentials once to access multiple applications.
A standard for transferring data between computers and peripheral devices.
Refers to the size of the drive bays. See also large form factor (LFF).
A specification for a transceiver for 10 Gigabit Ethernet technology.
A data storage device that uses solid-state memory to store persistent data. An SSD emulates a conventional hard disk drive. See also hard disk drive (HDD).
Describes a computer that does not require support from any other machine. Also called single system.
A low power mode that is activated on a server when AC power cords are connected and the service processor boots. In Standby power mode, power is supplied only to the components required to run the service processor. The operating system does not boot. See also Main power mode.
An advanced storage option used to deploy centrally pooled storage that appears to the operating system as local storage.
The combination of one or more physical disks into a single logical disk. The operating system views the logical disk as any other disk-based file system. See also logical disk and mirroring.
A designated user who has privileges to perform all administrative functions on a UNIX system. Also referred to as root. See also root.
A log that is accessible using Oracle ILOM. The syslog defines a set of common features for event logging and protocol for transmitting log entries to a remote host. This log can combine events form multiple Oracle ILOM sessions in a single place. Entries recorded in the syslog contain the same information contained in the local event log. See also event log.
A log that provides nonvolatile storage for system events that are logged autonomously by the service processor or directly with event messages sent from the host.
A log that is accessible using Oracle ILOM. This top-level presents a subset of relevant operational event log entries. The system log provides nonvolatile storage for system events that are logged autonomously by the service processor or directly with event messages from the host. See also event log and event.
In certain servers (such as the Oracle Servers X7-8 and X8-8), a module that contains the service processor, storage drive backplanes, real-time clock battery, energy storage module, internal host bus adapter, and USB port.
A window that emulates a particular type of terminal for running non-windowing programs. A terminal emulator is most commonly used for typing commands to interact with the computer's operating system.
A specified amount of time after which the server stops trying to finish a service routine that appears to be hung.
An event notification made by SNMP agents by their own initiative when certain conditions are detected. SNMP formally defines seven types of traps and permits subtypes to be defined. See also Simple Network Management Protocol (SNMP).
A feature that provides authenticity of a platform and its operating system. When enabled, TXT ensures that the operating system starts in a trusted environment, and provides the operating system with additional security capabilities not available to an untrusted operating system.
A standard for a dedicated microprocessor to secure hardware by using integrated cryptographic keys into devices.
A process used by Oracle Services field support personnel to repair and upgrade SP firmware when the SP flash image is corrupted. See also service processor (SP).
An Intel interconnect linking technology for communications between processors.
An error that cannot be corrected by error correction code (ECC). See also correctable error (CE).
Functionality that lets BIOS system variables, such as prioritized list of boot devices, to be viewed and manipulated by both the BIOS and Oracle ILOM for purposes of remote management and scale-out deployments. Enables Oracle ILOM to modify the BIOS Setup configuration and to preserve BIOS Setup configurations between BIOS upgrades.
A suite of tests that can be run manually or automatically from Oracle ILOM to test and detect problems on processor, memory, disk drives, and network ports.
A UEFI feature that defines how platform firmware can authenticate a digitally signed UEFI image, such as an operating system loader or a UEFI driver.
A characteristic of a signal that terminates an action. See also asserted.
A BIOS-compatible firmware interface that controls the system from power-on until an operating system is booted. UEFI allows you to configure, enable, disable, and manage system components. UEFI replaces Legacy BIOS in newer servers. See also Basic Input/Output System (BIOS).
A unique string that identifies a resource on the Internet or an intranet.
A device that provides emergency power to a system almost immediately if main power fails.
A 128-bit value used in distributed systems to uniquely identify information without requiring central coordination.
The process to remove access to a directory on a disk that is attached to a system or to a remote disk on a system. See also mount.
A record of essential user information that is stored on the system. Each user who accesses a system has a user account.
The number assigned to each user accessing a UNIX system. The system uses UID numbers to identify, by number, the owners of files and directories.
An attribute of a user account that designates the operations a user can perform and the resources a user can access.
An abstract specification for a computing device that can be implemented in different ways, in software or hardware.
Platform virtualization that enables multiple operating systems and applications to run in independent partitions, which behave like a virtual machine.
A virtual disk in which a filesystem, a database management system, or an application can place data.
A reboot or restart of the server. It requires cycling the server power from Main power mode to Standby power mode and back to Main power mode. See also cold reset.
A description of components that can be serviced while the server is in Standby power mode. See also Standby power mode.