Support | Mainframe Dictionary | R
A glossary of terms important to IBM mainframe machines
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z misc.
If you don't find what you seek,
please let us know.
R/390: Hybrid RS/6000 cum mainframe system announced in April 1996. Consists of an RS/6000 with an integrated mainframe processor card. Provides full mainframe and Unix capabilities, and supports AIX and all mainframe operating systems and applications. Obsolete. See also P/390, P/370, ptx.
RACF: Resource Access Control Facility. Security and access control system under z/OS and z/VM. The z/OS version is a component of SecureWay Security Server, but is also available as a stand-alone product. RACF authorizes access to resources (datasets, mini-disks, TSO/CMS/CICS/IMS TM log-on, etc) and logs unauthorized access attempts and accesses to protected datasets.
RACF/DB2 external security module: A RACF exit point that receives control from the DB2 access control authorization exit point (DSNX@XAC) to handle DB2 authorization checks. See also authorization checking.
RACF-indicated: An attribute of a dataset1 with the RACF indicator on. The dataset can only be used if RACF is installed and either a RACF profile or entry in the global access checking table exists for the dataset.
RACF remove ID utility: A program supplied with RACF that generates RACF commands to remove all references to deleted user IDs and group names. Alternatively, it can generate commands to remove all references to a specified user ID or group name.
Rack mount: A space efficient method of installing computer and electronics hardware. Rather than sitting on the floor or loosely on a shelf, a specially designed case fits into, and is securely fastened in, a standard sized shelving unit known as a rack.
RAD: Rapid Application Development. A generic term, which became fashionable in 1992, for a way of developing applications quickly. There’s no formal definition of RAD, but most RAD projects are built around the use of iterative or evolutionary prototyping using a 4GL. They also tend to involve SWAT teams to produce the code, JAD/JRP sessions to identify the real user requirements, and time-boxing as a pervasive management discipline.
Radio Shack: Arguably the first manufacturer to sell really big numbers of personal computers to consumers. Unfortunately, to reach a price point where people would buy, their early models were pretty awful, gaining the nickname Trash 80 for the TRS-80. For example, keyboard bounce was so bad that the keyboard device driver had code in it to remove two or more occurrences of the same character in quick succession. Buy, hey, that was 1977. The TRS in the TRS-80 is Tandy Radio Shack, reflecting the fact that Tandy had purchased Radio Shack in 1963.
RAID: Set of redundancy standards for disk subsystems (RAID 0-6), developed by the University of Berkeley and adopted by the RAID Advisory Board. RAID is set to be superseded by a more up-to-date set of standards (FRDS, FTDS, and DTDS). See also RAID 0, RAID 1, RAID 2, RAID 3, RAID 4, RAID 5, RAID 6, RAID 7.
RAID 0: Redundant Array of Inexpensive Disks – non-redundant striped array. Technically there is no redundancy at this level, but it does provide for speed advantages, compared with a single disk drive, by striping data in parallel sectors across multiple disk drives. The I/O transfer speed is increased for this architecture, however, a single drive element failure can result in an irrecoverable data loss. See also RAID.
RAID 1: Redundant Array of Inexpensive Disks – mirrored array. Independent data paths allow for complete disk duplication or data mirroring in this architecture. This level introduces redundancy in the sense that there are two copies of all data; this complete duplication also doubles the cost per megabyte. The speed of transfers is faster than a single drive because of the overlapping of reads and parallel writes. Access time can also be improved through accessing either copy of the data. Additional costs also derive from custom controllers and/or operating system changes. See also RAID.
RAID 2: Redundant Array of Inexpensive Disks – parallel array with ECC. Level 2 introduces hamming code error checking across the disks. This introduces the possibility of data recovery without a complete duplication of data, although it does require several check disks. This also requires that all disks in a group be accessed, even for small transfers, and the process has to wait for the slowest to finish before the transfer is complete. See also RAID.
RAID 3: Redundant Array of Inexpensive Disks – parallel array with parity. The distinguishing feature of level 3 is a single parity drive to accomplish redundancy. This is achieved by interleaving the parity information at the byte level. Typically, the drive spindles are synchronized. It still requires that all disks in a group be accessed, even for small transfers, with the wait for the slowest to finish before the transfer is complete. Spindle synchronization is expensive and often limits the choice of disk elements. See also RAID.
RAID 4: Redundant Array of Inexpensive Disks – striped array with parity. Level 4 introduces the concept of interleaving parity at the sector or transfer level. This permits faster individual disk reads for small transfers, and writes require accessing two disks. The parity check disk becomes a throughput bottleneck. See also RAID.
RAID 5: Redundant Array of Inexpensive Disks – striped array with rotating parity. Parity information is spiraled across all data drives in level 5, which attacks the problem of the parity disk bottleneck. This distributed parity increases write performance, but introduces high overhead to track the location of parity addresses. See also RAID.
RAID 6: Redundant Array of Inexpensive Disks. Level 6 provides a dedicated, cached, asynchronous parity drive. The single architectural difference between levels 4 and 5 is that level 5 distributes parity check information over all the disks in lieu of relegating it to a single disk. With level 6, the dedicated parity drive enjoys independent data and control paths, and can receive cached transfers via an independent asynchronous parity bus, a cache bus, or an external bus. Level 6 architecture should extend the dedicated cache, and the independent and asynchronous control/data paths to each disk in the array. See also RAID.
RAID 7: From Storage Computer Corporation (i.e., not a public RAID standard). Provides complete I/O independence and asynchrony. This is achieved by each I/O device/interface – including host interface device(s) – having a separate device cache, as well as independent control and data paths. Each device/interface is connected to a high speed data bus which has a central cache store capable of supporting multiple host I/O paths. A real time process oriented operating system is embedded into this disk array architecture. See also RAID.
RAIL: Redundant Arrays of Intelligent Libraries. An example of tape arrays exploiting the RAID concept. Tape arrays allow enhanced data reconstruction and enhanced integrity for archival data. Tape arrays may consist of a single automated tape library with data striped across connected drives to provide parallel data transfer across the drives. Another implementation builds on this concept by striping data across drives and tape libraries to permit tape mounting to occur in parallel. A third approach allows data to be striped on the tape and mounted on manual (non-automated) drives. See RAIT.
Raised floor: The machine room (also known as the Glasshouse). So named for the fact that you are walk on tiles that can be removed to gain access to the cabling that runs below your feet. The area below your feet is also the air conditioning duct to simplify cooling of all the hardware in the room. The irony of it all is that most raised floor machine rooms these days are built for ISP and corporate LAN servers, run by the same people who decried them during the PC Revolution.
RAM: Random Access Memory, also known as system memory, is that amount of physical memory that is addressable by and directly accessible to the processor chips on the motherboard or, much less frequently these days, on an add-on board on the bus.
RAMAC1: The IBM 350 RAMAC (vintage 1956) was the world’s first computer disk storage system.
RAMAC2: Mainframe RAID announced in June 1994. It’s a RAID 5 implementation, originally built around the Allicat 3.5 inch disks, and using 3380, 3390, or 9340 formats. Supports the 3990 extended functions (fast write, dual copy, concurrent copy, etc) as well as offering RAID 5 fault-tolerance. Available in two versions: the RAMAC Array DASD (9391), which attaches to a 3990, and the RAMAC Array Subsystem (9394/9395) with its own integrated controller, which attaches directly to channels. Promoted primarily on the basis of very high availability, rather than performance. Higher-capacity models (RAMAC 2) built around the 4GB UltraStar drives were announced in June 1995, with more RAID 5 features, and extended Remote copy. March 1996 ESCON connection and 3380-K format (four volumes per drawer) were added. September 1996, RAMAC 3 (9390) was introduced, based on the Ultrastar 9.1GB drive. Replaced by the 2105 August 2000.
RAMP-C: Registered Approach for Measurement Performance-COBOL, or Rochester [an IBM lab] Approximation of Machine Performance-COBOL (official IBM sources use both expansions of RAMP-C). An IBM commercial transaction benchmark. Written in COBOL, RAMP-C is probably a reasonable characterization of a small business machine workload, but not appropriate for database, ad hoc enquiries, etc. Comprises four types of interactive transaction from simple (equivalent to 70 lines of COBOL) to very complex (equivalent to 625 lines of COBOL). RAMP-C has attracted much criticism because IBM refuses to release the source code (although it does release a description of what the source code does), and competitors are unable to run it on their own machines (although many claim that they do). As a partial answer to the barrage of complaints about RAMP-C, IBM has had an independent audit done which seems to confirm many of IBM’s claims.
Rapid Network Reconnect: RNR utilizes the facilities of VTAM’s Multinode Persistent Session Services (MNPS) to improve system availability by allowing IMS TM to automatically reconnect terminal sessions following any kind of IMS1 failure and subsequent restart. RNR reduces network reconnect time after an IMS, z/OS, VTAM or CPC failure in a Sysplex environment.
RAS: Reliability, Availability, Serviceability. Highly desirable attributes of computer systems. IBM uses RAS as the touchstone of technical excellence for mainframe systems.
Raster: A predetermined pattern of lines that provides uniform coverage of a display area.
RAWT: Remote Abstract Windowing Toolkit. An implementation of Abstract Windowing Toolkit (AWT) for Java that allows Java applications on a z/OS or OS/400 host to display GUI data to and receive GUI data from a remote workstation. Java applications display and interact with GUI interfaces by using AWT APIs which are part of Sun’s JDK class library.
RC4: A royalty-based, RSA encryption algorithm (originally developed by a company called Rivest) that is widely used with SSL for Web-to-host security. For example, it’s one of the encryption algorithms supported by Host Access Client Package.
RCDS: Recovery Control Dataset.
RCF: Remote Console Facility.
RCMF/XRC: Remote Copy Management Facility/Extended Remote Copy. Automation code manages the remote copy configuration and storage subsystem(s), and provides protection against loss of data due to primary site outages. See also Remote copy, XRC.
RDBMS: Relational DataBase Management System. Generic term. DB2 is IBM’s RDBMS for almost every conceivable platform. Previously, IBM offered the Oracle RDBMS for the RT PC and RS/6000, and the Ingres RDBMS for the RT PC, RS/6000, and PS/2.
RDS: Remote Data Services. OS/2 facility which uses the LU6.2-based Communications Manager in OS/2 to give PC users access to data on other PCs across a local area network (LAN). Works by intercepting SQL database calls (not just data requests) and directing them to the appropriate networked PC.
Record Purpose Only: A change to the record of what is installed on a system, for IBM’s records only. No charges are involved.
Redbook: IBM manuals written less formally, by the ITSO, detailing a project they performed, including sample source code. More recently, redbooks have appeared that rival the text books you would find in a computer bookstore.
ReDiscovery: OS/2 and z/OS software, announced October 1993, which enables a user to create catalogs of bits of code. The software will seek out suitable code on the OS/2 workstation, in z/OS datasets, and in some z/OS libraries. IBM presents it as an infrastructure which will help organizations to reuse code. Replaced by SearchManager in September 1996.
Referential Integrity: The ability of relational databases to automatically ensure certain types of consistency within the database, e.g., when a row is deleted in one table, all related rows in all other tables are also deleted. Supported in DB2.
Refresh: IBMspeak for a release of software which incorporates corrective maintenance and other fixes needed since the previous release.
Registered network ID: An 8-byte name included in an IBM-maintained registry that is assigned to a particular IBM customer to identify a specific network.
Re-hosting: The frightfully trendy name that replaced downsizing, as people realized that downsizing was no longer the fashionable thing to be seen doing – my dear, I wouldn’t be caught dead downsizing – it’s just too, too vieux jeu. Re-hosting involves getting rid of a big mainframe and replacing it with one or more centralized machines, usually minicomputers.
Rejuvenation: Revitalizing host applications by providing them with a contemporary point-and-click graphical user interface (GUI) in place of their original, harsh, mainly textual, green-on-black screens.
Relational database: A type of database which allows information in one set of database tables to be connected to information in another set of tables without requiring duplication of information. See DB2.
Remote Copy: Facility introduced on the 3990-6 in March 1994. Links data between primary and secondary sites (up to 40km apart), reducing the time required to recover from loss of the primary system – data volumes can be switched over from the remote copy and do not have to be restored to DASD from tape. Comes in two forms: peer-to-peer remote copy between two ESCON-connected systems; and extended remote copy (XRC) across a telephone link. Extended remote copy uses CNT’s CHANNELink system. See also RCMF/XRC.
Remote Procedure Call: Remote Procedure Call (RPC) is a generic term for a procedure call between two systems. Although the two systems may be on the same machine, RPC usually refers to procedure calls across a network – one machine makes a procedure call, and the system bundles the call together with all the necessary communications code and sends it to another machine on the network for execution. The idea is to provide a simple mechanism for creating cooperative processing applications. There is no true RPC mechanism within SNA, although APPC provides all the technical infrastructure to build one. The IBM-backed OSF has attempted to create a standard for RPCs in the Unix environment, and it’s distinctly possible that this will be used as a model for other standards. See also APPC/MVS.
Remote Service Capability: Provides remote monitoring of system capacity, usage, and performance. IBM server or operating system fixes can be downloaded.
Remote Unit Of Work: A remote unit of work is a group of SQL statements which would need to be rolled back as a group if any single statement in the group could not be executed. The RUOW is the basic recovery unit where SQL originated on one machine accesses a database on another machine; for this reason, the RUOW is a critical function within a distributed database. Note that the RUOW concept allows updating on one other machine only (cf. Distributed Unit Of Work). Supported in DB2 and QMF. See also Atomic transaction.
Rent: Most of IBM’s business up to the mid 1960s was rental. Thereafter you had to buy your boxes. However, beginning in the early 1980s IBM started to return to a mainly rent policy, and now, via IBM Global Financing, is very strongly committed to rent (albeit under the guise of leasing).
Repository: Generic term for a generalized data dictionary cum database system which acts as a central store for information about systems. IBM leaked stories about its own repository for years but the project ran up against apparently endless problems. The first announcement of a repository product came in September 1989 with the late unlamented Repository Manager/MVS, and was followed by SystemView (September 1990) which also has its own repository (see Resource Object Data Manager). IBM still soldiers on with the concept, most recently with ObjectStore and TeamConnection.
Repository Manager/MVS: The very first tentative step (announced September 1989) towards a repository for application developers. Ran under DB2 in MVS/XA and ESA environments, and was intended to serve as the focal point for all shared application development within the AD/Cycle environment. Notwithstanding loads of announcements and marketing hype, Repository Manager/MVS was not an overnight success – more of an overnight disaster. By mid 1992 with a user base of just 30, IBM functionally stabilized it, not formally withdrawing it until July 1998.
Research: IBM Research has more than 3000 scientists and engineers at eight labs in six countries worldwide: the Thomas J Watson Research Center in Yorktown Heights, New York; the Almaden Research Center in San Jose, California; the Zurich Research Laboratory in Ruschlikon, Switzerland; the Tokyo Research Laboratory in Yamato, Japan; the Haifa Research Laboratory in Haifa, Israel; the China Research Laboratory in Beijing, China; the India Research Laboratory in Delhi, India; and the Austin Research Laboratory in Austin, Texas. The major areas of research are computer systems, computer applications and solutions, systems technology, physical sciences, mathematical sciences, data storage, and communications.
Resource Object Data Manager: The software (announced September 1991) providing Tivoli NetView for z/OS’s underlying repository of information. RODM provides the real-time data for the NetView Graphics Monitor Facility (see GMF). Also used to support automation. See also MIB, Repository.
Resource Reservation Protocol: An OS/400 Network Quality of Service (QoS) function for TCP/IP traffic. Includes an iSeries RSVP agent and X/Open standard APIs for applications. RSVP flow begins with the sender transmitting a PATH test to receiver; receiver transmits RESV message to sender; resources are reserved in routers along the path; data follows same route as PATH and RESV messages.
RESTRICTED attribute: A RACF user attribute that prevents the user ID from gaining access to protected resources that would otherwise be accessible through global access checking, UACC or an ID(*) entry in the access list.
RETAIN: REmote Technical Assistance and Information Network. System allowing customers and IBM access to a database of problems and fixes on IBM kit. RETAIN’s SSF (Service Support Facility) enables iSeries 400 systems to report problems and get fixes without user intervention; SSF is also available for other platforms.
Revenue bid: A bid (also known as an enterprise bid) based on a commitment by the customer to spend a particular amount of money with IBM over a particular period. The technique has recently found favor with IBM, and it can be quite a good deal for the customer, since IBM often throws in OSTAs and discounts as part of the deal. See also BEST.
REX: Route EXtension.
REXX: Restructured EXtended eXecutor language. A command procedure cum programming language which was initially available on z/VM only, replacing EXEC and EXEC2, but later became an SAA standard. Although SAA is a distant memory, REXX availability in z/OS TSO has seen it replace CLIST as the tool of choice, especially given the fact that TSO, and therefore REXX, can be run in batch. REXX is an effective programming language in its own right with powerful string processing facilities and is used to drive certain program products, notably GDDM. REXX is also available in VSE/ESA, AIX1, OS/2, Linux and Windows. Although normally interpretive, a REXX compiler and library is available for z/OS and z/VM. REXX for CICS is available for z/OS and VSE/ESA; it consists of REXX Development System and REXX Development System. Object REXX is available for Windows, OS/2, AIX, and Linux for Intel and zSeries 900. See also REXX Language Support.
RFC: Request For Comment. The process used by the IETF to evaluate and formulate internetworking related standards. Now the de facto standard scheme for Internet, intranet and multiprotocol networking.
RFC 1490: The standard that specifies how multiprotocol traffic can be encapsulated within Frame Relay. Often used as a short-hand term to refer native (i.e., minimum overhead) encapsulation of SNA/APPN within Frame Relay, even though the SNA/APPN encapsulation is defined by an addendum to this RFC known as the Frame Relay Forum FRF.3.1 specification.
RFI: Radio Frequency Interference. aka EMI. A measure of how much electrical noise is generated, and on what frequencies, by any kind of electronic equipment, not just computers. In the US, the FCC requires a sticker indicating their approval for all electronic equipment, based on standard tests for RFI.
RIC: Registered IBM Confidential. RICs are top-secret IBM documents; individually numbered copies of RIC documents are kept locked up in safes. Should such a document find its way out of IBM, its custodian is in big, big trouble, and could find him/herself in a Cemetery, Parking lot, Penalty box, or Cooling house.
RIFF: Resource Interchange File Format. Standard for storage of multimedia data, including audio, photo-quality images, and animation.
RIPPS: Retail Industry Programming Support Services.
RISC: Reduced Instruction Set Computer. Computer architecture using a very small set of instructions at the hardware level. The idea is to find the small set of instructions that are used most frequently, and then to design a chip that implements just these instructions, and executes them extremely quickly. More complex instructions are implemented in firmware and/or compilers. IBM invented the RISC concept, first used it in the RT PC, and then in the RS/6000. It also licenses the technology to third parties, including Sun Microsystems, which makes the Sparc RISC chip. RISCs are also used as auxiliary (channel, firmware) processors in the mainframe. There are good reasons to believe that the ultimate (literally) IBM architecture will be a parallel CEC using from one to thousands of RISC chips depending on how big a machine you want. Don’t believe that reduced means fewer – the RISC chip used in the RS/6000 has more instructions than the 370 had. See also VLIW.
RISC System/6000: RISC System/6000 is the official IBM name for what everybody else in the world calls the RS/6000. Rumor has it that Tandy has the right to the RS name, as in its Radio Shack subsidiary, and that IBM is too mean to buy it.
Risk management: Risk analysis, risk assessment and actually dealing with the threat, be it reactive, proactive or preventative action.
RLC: Recurring License Charge.
RMDS: Report Management and Distribution System. Tivoli z/OS JESx system for managing printer output. Includes report distribution facilities, such as bundling and selective printing, and viewing facilities such as scroll back and search. Terminal access can be through VTAM, TSO, CICS and IMS TM, security is provided through RACF, and the package includes an on-line help facility. Can be used to manage SYSOUT under VSAM and QSAM. Originally developed for internal use within IBM, but foisted on an unenthusiastic user base, faute de mieux.
RMI: Remote Method Invocation. Distributed object invocation that enables Java applications to remotely invoke the methods of remote Java objects even if they are on a different host. RMI can run over either JRMP or IIOP. The Java equivalent of RPC.
ROI: Return On Investment.
ROJ: Retired On the Job. IBMspeak for an employee who is just marking time rather than doing his job.
ROLAP: Relational On-Line Analytical Processing.
Rolm: Company specializing in telephone and voice equipment. Taken over by IBM in 1984. IBM had trouble getting Rolm to perform (Rolm was rumored to be costing IBM $100M per year), and in December 1988 IBM set up a 50/50 company with Siemens to sell Rolm products in the US and sold the manufacturing and product development of Rolm kit to Siemens. IBM sold out completely to Siemens in May 1992.
ROR: Resource Owning Region.
Router: A generic term for a device for connecting two networks which makes interconnected networks (usually LANs) appear as a single network to attached devices. A router performs a similar function to a bridge, but functions at the network layer (layer 3 of the OSI model), and routes traffic using a network address. Routers generally support a much wider range of network management functions than bridges, including load balancing, partitioning of the network, statistics collection, diagnostics and repair, etc. Routers are more appropriate than bridges for large, inter-enterprise networks with high security requirements. IBM’s first effort was the 6611, announced in January 1992. See also 2210.
RouteXpander/2: PS/2-based router gateway for smallish locations, announced September 1992. Provides Frame Relay support, and can act as a feeder to a 6611 router or 3745 communications processor. Replaced in March 1994 by a family of enhanced RouteXpander/2 products and the X.25 Xpander/2. Withdrawn July 1998, though support had already ended in December 1995.
Routing Information Field: A field in the Layer-2 802.5 MAC1 header used by Source Route Bridging (SRB) to specify the LAN-to-LAN route that should be traversed by packets forwarded from one LAN to another.
RP3: Research Parallel Processor Project. Early IBM research project to build a parallel CEC using RISC engines (up to 512 at once) and the Mach operating system. The system went live in September 1988.
RPG: Report Program Generator. Programming language widely used on the eserver iSeries 400 and its predecessors: AS/400 and System/3x. Coding is column-dependent fill-in-the-blanks forms. As its name implies, best used for report generation, and very strong sort/merge facilities, where it can be used completely non-procedurally. IBM created RPG in 1965. RPG/370 was the mainframe SAA CPI version announced February 1991 and withdrawn August 1993. It ran on z/OS and z/VM. For VSE/ESA, there is DOS/VS RPG II, still available, but it has been more than two decades since the last release. OS/VS RPG II, for z/OS, is even older. Hopefully, someone checked them both for Y2K compatibility. RPG’s major popularity on the mainframe came at the low end, as it was the only programming language that ran very well on the System/360 Model 20. For the iSeries 400, where virtually all of the RPG use occurs today, WebSphere Development Studio for iSeries includes ILE RPG (RPG IV), VisualAge RPG and CODE/400, the tools of choice for RPG programmers. They run on OS/400 and Windows.
RPL1: Remote Program Load.
RPL2: Request Parameter List. A VTAM control block that contains the parameters necessary for processing a request for data transfer. Likewise for VSAM, both the Assembler macro1 and the control block for an I/O request.
RPS2: Rotational Position Sensing. A hardware technique for allegedly increasing disk performance by connecting to the disk only during the period when the appropriate part of the disk is under the read/write head; that way the channel can be freed to do other work instead of having to wait for the disk to rotate. First used by IBM in the 3330 DASD.
RRAS1: Remote Relational Access Support. A z/VM facility that was built on TSAF for distributed data management. Enabled databases to be marked as shared system resources so that DB2 data could be accessed by a remote DB2 machine.
RRDS: Relative Record DataSet. VSAM file type – can be used as a replacement for BDAM. Each record can be accessed directly by its record number.
RR packet: Receive ready packet. A packet used by a data terminal equipment (DTE) or by a data circuit-terminating equipment (DCE2) to indicate that it is ready to receive data packets within the window.
RS/6000: IBM’s principal challenger in the Unix workstation and mid-range market, which finally emerged in February 1990 after a long, difficult, and delayed birth. The initial IBM posture was that the box was to be sold into the engineering and scientific market, but subsequently it turned into a generic Unix machine for all markets. October 1994 symmetric multiprocessing versions were announced. Replaced by the eserver pSeries October 3, 2000, which also absorbed the NUMA-Q acquired through an IBM merger with Stratus. Runs IBM’s Unix derivative: AIX1. See Pegasus1.
RS-232-C: An ANSI interface standard. The most commonly applied standard for the interconnection of asynchronous devices. RS-232-C uses the ubiquitous 25-pin D-plug/socket with ribbon cable, although you may come across pins/sockets with only 9 pins. There is a closely defined specification of the functions of each pin, including voltage levels etc, although for most applications four or five pins are sufficient. Used by the mini and micro manufacturers for local connection of screens and (serially) printers. Also used by robots and programmable controllers. Even IBM has been known to use it. RS-232-C is roughly equivalent to ITU-T V24.
RSA: A commercial public key encryption technology, owned by RSA Security Inc. PGP another common public key encryption technology is also patented by RSA, hence it cannot be used in commercial applications.
RSA Security: Company, specializing in cryptographic software, with which IBM set up a venture in January 1996 to jointly develop products for secure electronic commerce.
RSCS: Remote Spooling Communications Subsystem. A multi-host networking file transfer system, similar in some respects to JES2 Remote Telecommunications Access Method (RTAM). Allows users on one z/VM system to send messages, files, commands, and jobs to users, workstations, and printers on other systems. Runs under z/VM and is compatible with z/OS JESx subsystems and VSE/ESA POWER1. Supports remote printers and readers, and links to other RTAMs for z/VM. Based on an IBM internal product called VNET. Supports VTAM with CPU-to-CPU working.
RSM: Real Storage Manager. The part of z/OS which controls real memory.
RST: Remote modem Self-Test. A remote check on hardware in the field to identify whether a unit is failing.
RTG: Routing Table Generator. IBM software that generates path tables for SNA networks. Announced June 1985. Support ended August 1991. Withdrawn December 1997 because of Y2K concerns. See also NETDA/2.
RTM1: Response Time Monitor. An optional hardware facility on some IBM terminal control units that collects ranges of response times for each device, and can transmit the data to the host – normally to NLDM.
RTM2: Recovery Termination Manager. z/OS component which controls normal and abnormal task completion. The application programmer can supply his own exits to process certain error conditions rather than letting the operating system do it.
RT PC: RISC Technology Personal Computer. The multi-user workstation, announced early 1986. Used a 32-bit RISC processor, running under AIX. Was never a great success (it’s estimated that the RT PC took under 2% of the worldwide Unix market) and was replaced by the RS/6000 in February 1990. See also 6150, 6151, 6152.
Rusty memory: Industry speak for data storage that uses iron-oxide-based magnetic media such as tape, as well as antique removable disk packs.
RXSQL: Interpreter Interface to SQL/DS. z/VM IBM program offering which provides an interface between REXX and DB2 (formerly known as SQL/DS in z/VM). Withdrawn December 1997 though support had ended May 1988.