Multiple Virtual Storage, more commonly called MVS, was the most commonly used operating system on the System/370 and System/390 IBM mainframe computers. It was developed by IBM, but is unrelated to IBM's other mainframe operating system, VM.
First released in 1974, MVS had been renamed multiple times, first to MVS/XA (eXtended Architecture), next to MVS/ESA (Enterprise Systems Architecture), then to OS/390 (when UNIX System Services (USS) were added), and finally to z/OS (when 64-bit support was added with the zSeries models). Its core remains fundamentally the same operating system. By design, programs written for MVS can still run on z/OS without modification.
At first IBM described MVS as simply a new release of OS/VS2. But it was in fact a complete re-write - previous OS/VS2 releases were upgrades of OS/MVT and, like MVT, were mainly written in Assembler; the core of MVS was almost entirely written in PL/S. IBM's use of "OS/VS2" emphasized upwards compatibility: application programs which ran under MVT did not even need to be re-compiled in order to run under MVS; the same Job Control Language files could be used unchanged; the utilities and other non-core facilities like TSO ran unchanged. But users almost unanimously called the new system MVS from the start, and IBM followed their lead in the naming of later major versions such as MVS/XA. After the release of MVS, users described earlier OS/VS2 releases as SVS (Single Virtual Storage).
OS/MVT (Multitasking with a Variable number of Tasks) was an enhancement which further refined memory usage. Instead of using fixed-size memory partitions, MVT allocated memory to programs as needed provided there was enough contiguous physical memory available. This was a significant advance over MFT's memory management: there was no predefined limit on the number of jobs that could run at the same time; and two or more large jobs could run at the same time if enough memory was available. But it had some weaknesses: if a job allocated memory dynamically (as most sort programs and database management systems do), the programmers had to estimate the job's maximum memory requirement and pre-define it for MVT; a job which contained a mixture of small and large programs would waste memory while the small programs were running; most seriously, memory could become fragmented, i.e. the memory not used by current jobs could be divided into uselessly small chunks between the areas used by current jobs, and the only remedy was to wait until all current jobs finished before starting any new ones.
In the early 1970s IBM sought to mitigate these difficulties by introducing virtual memory (referred to by IBM as "virtual storage"), which allowed programs to request address spaces larger than physical memory. The original implementations had a single virtual address space, shared by all jobs. OS/VS1 was OS/MFT within a single virtual address space; OS/VS2 SVS was OS/MVT within a single virtual address space. So OS/VS1 and SVS in principle had the same disadvantages as MFT and MVT but the impacts were less severe because jobs could request much larger address spaces.
| MVS address spaces - global view
| One application's view
MVS originally supported 24-bit addressing (i.e. up to 16MB). As the underlying hardware progressed it supported 31-bit (XA and ESA; up to 2048MB) and now (as z/OS) 64-bit addressing. Two of the most significant reasons for the rapid upgrade to 31-bit addressing were: the growth of large transaction-processing networks, mostly controlled by CICS, which ran in a single address space; the DB2 relational database management system needed more than 8MB of application address space in order to run efficiently (early versions were configured into two address spaces which communicated via the shared virtual area, but this imposed a significant overhead since all such communications had to be transmitted via the operating system).
The main user interfaces to MVS are: Job Control Language (JCL), which was originally designed for batch processing but from the 1970s onwards was also used to start and allocate resources to long-running interactive jobs such CICS; and TSO (Time Sharing Option), the interactive time-sharing interface, which was mainly used to run development tools and a few end-user information systems. ISPF is a TSO application for users on 3270-family terminals (and later, on VM as well) which allows the user to accomplish the same tasks as TSO's command line but in a menu and form oriented manner, and with a full screen editor and file browser. TSO's basic interface is command line, although facilities were added later for creating form-driven interfaces).
Early editions of MVS (mid-1970s) were among the first of the IBM OS series to support multiprocessor configurations, though it had previously been supported in the 1960s on a limited basis by the M65MP variant of OS/360 running on 360/65 and 360-67. The 360-67 had also hosted the multiprocessor capable TSS/360 and MTS operating systems. In tightly-coupled systems 2 (later up to 8) CPUs shared concurrent access to the same memory (and copy of the operating system) and peripherals, providing greater processing power and a degree of graceful degradation if one CPU failed. In loosely-coupled configurations each of a group of processors (single and / or tightly-coupled) had its own memory and operating system but shared peripherals and the operating system component JES3 allowed the whole group to be managed from one console - this provided greater resilience and enabled operators to decide which processor should run which jobs from a central job queue.
Multiple copies of MVS (or other IBM operating systems) could share the same machine if that machine was controlled by VM/370 - in this case VM/370 was the real operating system and regarded the "guest" operating systems as applications with unusually high privileges. As a result of later hardware enhancements one instance of an operating system (either MVS, or VM with guests, or other) could also occupy a Logical Partition (LPAR) instead of an entire physical system.
Multiple MVS instances can be organized and collectively administered in a structure called a systems complex or sysplex, introduced in September, 1990. Instances interoperate through a software component called a Cross-system Coupling Facility (XCF) and a hardware component called a Hardware Coupling Facility (CF or Integrated Coupling Facility, ICF, if co-located on the same mainframe hardware). Multiple sysplexes can be joined via standard network protocols such as IBM's proprietary Systems Network Architecture (SNA) or, more recently, via TCP/IP.
The z/OS operating system (MVS' most recent descendant) also has native support to execute POSIX applications.
The native encoding scheme of IBM mainframes and their peripherals is Big Endian EBCDIC, but MVS provides hardware-accelerated services to perform translation and support of ASCII, Little Endian, and Unicode.
As with other members of the OS family, MVS' data sets were record-oriented. MVS inherited three main types from its predecessors:
Sequential and ISAM datasets could store either fixed-length or variable length records, and all types could occupy more than one disk volume.
All of these are based on the VTOC disk structure.
Early IBM database management systems used various combinations of ISAM and BDAM datasets - usually BDAM for the actual data storage and ISAM for indexes.
VSAM also included a catalog component which was used for MVS' master catalog.
Partitioned datasets (PDS) were sequential datasets which were subdivided into "members" which could be processed as sequential files in their own right. The most important use of PDS was for program libraries - system administrators used the main PDS as a way to allocate disk space to a project and the project team then created and edited the members.
Generation Data Groups (GDGs) were originally designed to support grandfather-father-son backup procedures - if a file was modified, the changed version became the new "son", the previous "son" became the "father", the previous "father" became the "grandfather" and the previous "grandfather" was deleted. But one could set up GDGs with a lot more than 3 generations and some applications used GDGs to collect data from several sources and feed the information to one program - each collecting program created a new generation of the file and the final program read the whole group as a single sequential file (by not specifying a generation in the JCL).
Modern versions of MVS (i.e. z/OS) also support POSIX-compatible "slash" filesystems along with facilities for integrating the two filesystems. That is, the OS can make an MVS dataset appear as a file to a POSIX program or subsystem. These newer filesystems include Hierarchical File System (HFS) (not to be confused with Apple's Hierarchical File System) and zFS (not to be confused with Sun's ZFS).
While z/OS continues to run older 24-bit MVS applications, it does so alongside just written 64-bit hardware-accelerated Java running with state-of-the-art security and Unicode XML data formats on protected filesystems with flexible mounting and long filenames, communicating across IPv6 Hipersocket virtual networks and/or high speed CFs to 64-bit geographically clustered relational databases that use the latest performance-enhancing SQL features including materialized query tables and cube views — for example.