Misplaced Pages

#941058

48-428: Early work on Sprite was based on the idea of making the operating system more "network aware", and thereby at the same time make it invisible to the user. The primary area of work was the building of a new network file system which made heavy use of local client-side caching in order to improve performance. After opening the file and some initial reads, the network is only used on-demand, and most user actions occur against

96-483: A basic memory cycle of 2.18 μs, using the IBM 7302 Core Storage core memory technology from the IBM 7030 (Stretch) project. With a processing speed of around 100 Kflop/s , the 7090 is six times faster than the 709, and could be rented for half the price. An upgraded version, the 7094, was up to twice as fast. Both the 7090 and the 7094 were withdrawn from sale on July 14, 1969, but systems remained in service for more than

144-492: A decade after. In 1961, the IBM 7094 famously employed a speech synthesis program to sing " Daisy Bell ", becoming something of a cultural icon . Although the 709 was a superior machine to its predecessor, the 704, it was being built and sold at the time that transistor circuitry was supplanting vacuum tube circuits. Hence, IBM redeployed its 709 engineering group to the design of a transistorized successor. That project became called

192-522: A less expensive IBM 7044 to handle input/output (I/O) with the 7094 performing mostly computation. Aerospace developed the Direct Couple operating system, an extension to IBSYS, which was shared with other IBM customers. IBM later introduced the DCS as a product. The 7090 used more than 50,000 germanium alloy-junction transistors and (faster) germanium diffused junction drift transistors . The 7090 used

240-542: A major technological shift in the history of computing. By allowing many users to interact concurrently with a single computer, time-sharing dramatically lowered the cost of providing computing capability, made it possible for individuals and organizations to use a computer without owning one, and promoted the interactive use of computers and the development of new interactive applications . The earliest computers were extremely expensive devices, and very slow in comparison to later models. Machines were typically dedicated to

288-655: A particular set of tasks and operated by control panels, the operator manually entering small programs via switches in order to load and run a series of programs. These programs might take hours to run. As computers grew in speed, run times dropped, and soon the time taken to start up the next program became a concern. Newer batch processing software and methodologies, including batch operating systems such as IBSYS (1960), decreased these "dead periods" by queuing up programs ready to run. Comparatively inexpensive card punch or paper tape writers were used by programmers to write their programs "offline". Programs were submitted to

336-599: A second deployment of CTSS was installed on an IBM 7094 that MIT has purchased using ARPA money. This was used to support Multics development at Project MAC . JOSS began time-sharing service in January 1964. Dartmouth Time-Sharing System (DTSS) began service in March 1964. Throughout the late 1960s and the 1970s, computer terminals were multiplexed onto large institutional mainframe computers ( centralized computing systems), which in many implementations sequentially polled

384-645: A wide variety of software provided for them by IBM. In addition, there was a very active user community within the user organization, SHARE . IBSYS is a "heavy duty" production operating system with numerous subsystem and language support options, among them FORTRAN , COBOL , SORT/MERGE, the MAP assembler, and others. FMS, the Fortran Monitor System , was a more lightweight but still very effective system optimized for batch FORTRAN and assembler programming. The assembler provided, FAP, ( FORTRAN Assembly Program ),

432-500: Is not zero, it selects just one of seven index registers, however, the program can return to multiple tag mode with the instruction Enter Multiple Tag Mode , restoring 7090 compatibility. In April 1964, the first 7094 II was installed, which had almost twice as much general speed as the 7094 due to a faster clock cycle , dual memory banks and improved overlap of instruction execution, an early instance of pipelined design. In 1963, IBM introduced two new, lower cost machines called

480-420: Is the concurrent sharing of a computing resource among many tasks or users by giving each task or user a small slice of processing time . This quick switch between tasks or users gives the illusion of simultaneous execution. It enables multi-tasking by a single user or enables multiple-user sessions. Developed during the 1960s, its emergence as the prominent model of computing in the 1970s represented

528-457: Is the fourth member of the IBM 700/7000 series scientific computers. The first 7090 installation was in December 1959. In 1960, a typical system sold for $ 2.9 million (equivalent to $ 23 million in 2023) or could be rented for $ 63,500 a month (equivalent to $ 501,000 in 2023). The 7090 uses a 36-bit word length , with an address space of 32,768 words (15-bit addresses). It operates with

SECTION 10

#1732885114942

576-715: Is time-sharing". For DEC, for a while the second largest computer company (after IBM), this was also true: Their PDP-10 and IBM's 360/67 were widely used by commercial timesharing services such as CompuServe, On-Line Systems, Inc. (OLS), Rapidata and Time Sharing Ltd. The advent of the personal computer marked the beginning of the decline of time-sharing. The economics were such that computer time went from being an expensive resource that had to be shared to being so cheap that computers could be left to sit idle for long periods in order to be available as needed. Although many time-sharing services simply closed, Rapidata held on, and became part of National Data Corporation . It

624-441: The IBM 1403 line printer. Later IBM introduced the 7094/7044 Direct Coupled System ; the 7044 handled spooling between its fast 1400-series peripherals and 1301 or 1302 disk files, and used data channel to data channel communication as the 7094's interface to spooled data, with the 7094 primarily performing computations. There is also a 7090/7040 DCS. The 7090 and 7094 machines were quite successful for their time, and had

672-516: The IBM 2741 ) with two different seven-bit codes. They would connect to the central computer by dial-up Bell 103A modem or acoustically coupled modems operating at 10–15 characters per second. Later terminals and modems supported 30–120 characters per second. The time-sharing system would provide a complete operating environment, including a variety of programming language processors, various software packages, file storage, bulk printing, and off-line storage. Users were charged rent for

720-546: The IBM 7040 and 7044. They have a 36-bit architecture based on the 7090, but with some instructions omitted or optional, and simplified input/output that allows the use of more modern, higher performance peripherals from the IBM 1400 series . The 7094/7044 Direct Coupled System (DCS) was initially developed by an IBM customer, the Aerospace Corporation , seeking greater cost efficiency and scheduling flexibility than IBM's IBSYS tape operating system provided. DCS used

768-465: The Standard Modular System (SMS) cards using current-mode logic some using diffused junction drift transistors. The basic instruction formats were the same as the IBM 709 : The documentation of opcodes used signed octal. The flag field indicated whether to use indirect addressing or not. The decrement field often contained an immediate operand to modify the results of the operation, or

816-537: The UK . By 1968, there were 32 such service bureaus serving the US National Institutes of Health (NIH) alone. The Auerbach Guide to Timesharing (1973) lists 125 different timesharing services using equipment from Burroughs , CDC , DEC , HP , Honeywell , IBM , RCA , Univac , and XDS . In 1975, acting president of Prime Computer Ben F. Robelen told stockholders that "The biggest end-user market currently

864-422: The 1970s, Ted Nelson 's original " Xanadu " hypertext repository was envisioned as such a service. Time-sharing was the first time that multiple processes , owned by different users, were running on a single machine, and these processes could interfere with one another. For example, one process might alter shared resources which another process relied on, such as a variable stored in memory. When only one user

912-417: The 709 and 7090, and requires a Leave Multiple Tag Mode instruction in order to enter seven index register mode and use all seven index registers. In multiple tag mode, when more than one bit is set in the tag field, the contents of the two or three selected index registers are logically ORed , not added , together, before the decrement takes place. In seven index register mode, if the three-bit tag field

960-496: The 709-T (for transistorized ), which because of the sound when spoken, quickly shifted to the nomenclature 7090 (i.e., seven - oh - ninety). Similarly, the related machines such as the 7070 and other 7000 series equipment were sometimes called by names of digit - digit - decade (e.g., seven - oh - seventy). An upgraded version, the IBM 7094 , was first installed in September 1962. It has seven index registers , instead of three on

1008-527: The University of Illinois in early 1961. Bitzer has long said that the PLATO project would have gotten the patent on time-sharing if only the University of Illinois had not lost the patent for two years. The first interactive , general-purpose time-sharing system usable for software development, Compatible Time-Sharing System , was initiated by John McCarthy at MIT writing a memo in 1959. Fernando J. Corbató led

SECTION 20

#1732885114942

1056-415: The cache. Similar utilities allow remote devices to be mapped into the local computer's space, allowing for network printing and similar duties. Many of the key Unix files are based on the network, including things like the password file. All machines in a network share the root directory as well. Other common Unix utilities such as finger were re-written to make them network aware as well, listing all of

1104-430: The change in the meaning of the term time-sharing a source of confusion and not what he meant when he wrote his paper in 1959. There are also examples of systems which provide multiple user consoles but only for specific applications, they are not general-purpose systems. These include SAGE (1958), SABRE (1960) and PLATO II (1961), created by Donald Bitzer at a public demonstration at Robert Allerton Park near

1152-705: The computer much the same way that the average household buys power and water from utility companies." Christopher Strachey , who became Oxford University's first professor of computation, filed a patent application in the United Kingdom for "time-sharing" in February 1959. He gave a paper "Time Sharing in Large Fast Computers" at the first UNESCO Information Processing Conference in Paris in June that year, where he passed

1200-440: The computer's resources, such as when a large JOSS application caused paging for all users. The JOSS Newsletter often asked users to reduce storage usage. Time-sharing was nonetheless an efficient way to share a large computer. As of 1972 DTSS supported more than 100 simultaneous users. Although more than 1,000 of the 19,503 jobs the system completed on "a particularly busy day" required ten seconds or more of computer time, DTSS

1248-528: The concept on to J. C. R. Licklider . This paper was credited by the MIT Computation Center in 1963 as "the first paper on time-shared computers". The meaning of the term time-sharing has shifted from its original usage. From 1949 to 1960, time-sharing was used to refer to multiprogramming without multiple user sessions. Later, it came to mean sharing a computer interactively among multiple users. In 1984 Christopher Strachey wrote he considered

1296-581: The concept, but did not use the term, in the 1954 summer session at MIT . Bob Bemer used the term time-sharing in his 1957 article "How to consider a computer" in Automatic Control Magazine and it was reported the same year he used the term time-sharing in a presentation. In a paper published in December 1958, W. F. Bauer wrote that "The computers would handle a number of problems concurrently. Organizations would have input-output equipment installed on their own premises and would buy time on

1344-409: The development of the system, a prototype of which had been produced and tested by November 1961. Philip M. Morse arranged for IBM to provide a series of their mainframe computers starting with the IBM 704 and then the IBM 709 product line IBM 7090 and IBM 7094 . IBM loaned those mainframes at no cost to MIT along with the staff to operate them and also provided hardware modifications mostly in

1392-434: The earlier machines. The 7151-2 Console Control Unit for the 7094 has a distinctive box on top that displays lights for the four new index registers. The 7094 introduced double-precision floating point and additional instructions , but largely maintained backward compatibility with the 7090. Although the 7094 has four more index registers than the 709 and 7090, at power-on time it is in multiple tag mode , compatible with

1440-438: The earliest days of personal computers, many were in fact used as particularly smart terminals for time-sharing systems. DTSS's creators wrote in 1968 that "any response time which averages more than 10 seconds destroys the illusion of having one's own computer". Conversely, timesharing users thought that their terminal was the computer, and unless they received a bill for using the service, rarely thought about how others shared

1488-444: The form of RPQs as prior customers had already commissioned the modifications. There were certain stipulations that governed MIT's use of the loaned IBM hardware. MIT could not charge for use of CTSS. MIT could only use the IBM computers for eight hours a day; another eight hours were available for other colleges and universities; IBM could use their computers for the remaining eight hours, although there were some exceptions. In 1963

Sprite (operating system) - Misplaced Pages Continue

1536-409: The operations team, which scheduled them to be run. Output (generally printed) was returned to the programmer. The complete process might take days, during which time the programmer might never see the computer. Stanford students made a short film humorously critiquing this situation. The alternative of allowing the user to operate the computer directly was generally far too expensive to consider. This

1584-511: The original starter to move it to another machine on the network, or take control of it locally again. Long tasks (like compiling the Sprite system) can appear very fast. Work on the "early" Sprite outlined above ended around 1987, but was improved during the next year. Starting in 1990 Sprite was used as the basis for development of the first log-structured file system (LFS), development of which continued until about 1992. LFS dramatically increases

1632-498: The people logged on across the network. This makes a Sprite network appear as if it were a single large time-sharing system, or a single-system image . Another key addition to Sprite is process migration , which allows programs to be moved between machines at any time. The system maintains a list of machines and their state, and automatically moves processes to idle machines to improve local performance. Processes can also be "evicted" from machines to improve their performance, causing

1680-523: The performance of file writes at the expense of read performance. Under Sprite, this tradeoff is particularly useful because most read access is cached anyway—that is, Sprite systems typically perform fewer reads than a normal Unix system. LFS-like systems also allow for much easier crash recovery, which became a major focus of the project during this period. Additional experimentation on striped file systems, both striped across different machines as well as clusters of drives, continued until about 1994. Sprite

1728-529: The rise of microcomputing in the early 1980s, time-sharing became less significant, because individual microprocessors were sufficiently inexpensive that a single person could have all the CPU time dedicated solely to their needs, even when idle. However, the Internet brought the general concept of time-sharing back into popularity. Expensive corporate server farms costing millions can host thousands of customers all sharing

1776-648: The same common resources. As with the early serial terminals, web sites operate primarily in bursts of activity followed by periods of idle time. This bursting nature permits the service to be used by many customers at once, usually with no perceptible communication delays, unless the servers start to get very busy. Genesis In the 1960s, several companies started providing time-sharing services as service bureaus . Early systems used Teletype Model 33 KSR or ASR or Teletype Model 35 KSR or ASR machines in ASCII environments, and IBM Selectric typewriter -based terminals (especially

1824-724: The terminal, a charge for hours of connect time, a charge for seconds of CPU time, and a charge for kilobyte-months of disk storage. Common systems used for time-sharing included the SDS 940 , the PDP-10 , the IBM 360 , and the GE-600 series . Companies providing this service included GE 's GEISCO , the IBM subsidiary The Service Bureau Corporation , Tymshare (founded in 1966), National CSS (founded in 1967 and bought by Dun & Bradstreet in 1979), Dial Data (bought by Tymshare in 1968), AL/COM , Bolt, Beranek, and Newman (BBN) and Time Sharing Ltd. in

1872-469: The terminals to see whether any additional data was available or action was requested by the computer user. Later technology in interconnections were interrupt driven, and some of these used parallel data transfer technologies such as the IEEE 488 standard. Generally, computer terminals were utilized on college properties in much the same places as desktop computers or personal computers are found today. In

1920-401: The time. Printing and punched card I/O, however, employed the same modified unit record equipment introduced with the 704 and was slow. It became common to use a less expensive IBM 1401 computer to read cards onto magnetic tape for transfer to the 7090/94. Output would be written onto tape and transferred to the 1401 for printing or card punching using its much faster peripherals, notably

1968-402: Was able to handle the jobs because 78% of jobs needed one second or less of computer time. About 75% of 3,197 users used their terminal for 30 minutes or less, during which they used less than four seconds of computer time. A football simulation, among early mainframe games written for DTSS, used less than two seconds of computer time during the 15 minutes of real time for playing the game. With

Sprite (operating system) - Misplaced Pages Continue

2016-751: Was an indirect effective address ; i.e., fetch the word at that location and treat the T and Y fields as described above. Data formats are Octal notation was used in documentation and programming; console displays lights and switches were grouped into three-bit fields for easy conversion to and from octal. The 7090 series features a data channel architecture for input and output, a forerunner of modern direct memory access I/O. Up to eight data channels can be attached, with up to ten IBM 729 tape drives attached to each channel. The data channels have their own very limited set of operations called commands. These are used with tape (and later, disk) storage as well as card units and printers, and offered high performance for

2064-429: Was because users might have long periods of entering code while the computer remained idle. This situation limited interactive development to those organizations that could afford to waste computing cycles: large universities for the most part. The concept is claimed to have been first described by Robert Dodds in a letter he wrote in 1949 although he did not use the term time-sharing . Later John Backus also described

2112-508: Was not a microkernel system, and suffers the same sort of problems as other Unixes in terms of development complexity, becoming increasingly difficult to develop as more functionality was added. By the 1990s it was suffering and the small team supporting the project was simply not able to keep up with the rapid changes in Unix taking place during this time. The project was slowly shut down by 1994. Time-sharing In computing , time-sharing

2160-419: Was primarily driven by the time-sharing industry and its customers. Time-sharing in the form of shell accounts has been considered a risk. Significant early timesharing systems: IBM 7090 The IBM 7090 is a second-generation transistorized version of the earlier IBM 709 vacuum tube mainframe computer that was designed for "large-scale scientific and technological applications". The 7090

2208-479: Was still of sufficient interest in 1982 to be the focus of "A User's Guide to Statistics Programs: The Rapidata Timesharing System". Even as revenue fell by 66% and National Data subsequently developed its own problems, attempts were made to keep this timesharing business going. Beginning in 1964, the Multics operating system was designed as a computing utility , modeled on the electrical or telephone utilities. In

2256-411: Was used to further define the instruction type. The tag field might describe an index register to be operated on, or be used as described below. The Y field might contain an address, an immediate operand or an opcode modifier. For instructions where the tag field indicated indexing, the operation was If there was no F field or F is not all one bits, then the above was the effective address . Otherwise it

2304-539: Was using the system, this would result in possibly wrong output - but with multiple users, this might mean that other users got to see information they were not meant to see. To prevent this from happening, an operating system needed to enforce a set of policies that determined which privileges each process had. For example, the operating system might deny access to a certain variable by a certain process. The first international conference on computer security in London in 1971

#941058