CSIRO Computing History, Chapter 3

By Robert C. Bell August 12th, 2019

 

Division of Computing Research and Csironet – growth. 1967-1980s

Last updated: 3 June 2020.
Robert C. Bell
Status: Commenced 1978-79.

previous chapterComputing History Homenext chapter

Contents

(under development)

1967 – DCR, DAD, Document Region management, disc
1968 – DCR, widening applications, displays, more branches, evolution of on-line storage and management, first PDP, interactive editing (FRED), teletypes, first remote access, Trevor Pearcey seconded to CDC
1968-69 – DCR, displays, appeal for expansion, Perth branch
1969-70 – DCR, wide-area network=CSIRONET, access to CDC 6600, multi-user disc access, interactive job scheduling, computer-to-computer link, visualisation
1970-71 – DCR, two-bank DAD, growth in on-line storage demands – management, Townsville office, job submission via telex, start of network project, interstate lines, PDP-11s, text processing, small computers, DAD exported
1971-72 – DCR, CSIRONET in operation, PDP-11 nodes at branches, Rockhampton branch, virtual Document Region, reentrant FRED, improved reliability, Claringbold acting chief

1972-73 – DCR, Cyber 76 announced and arrives, Margaret Thatcher visits, DR retention periods and becomes and HSM, extension of CSIRONET to provide local nodes, TED replaces FRED – boxes

1973-74 – DCR, Cyber 76 in service: links, porting apps, permanent file management; CSIRONET: 30 nodes, delays in lines for network, Remote Access Group; AI, 3200s removed, building extension announced

1974-75 – DCR, COMp80 in service: microfiche, image processing, Hobart branch, Cyber 76 low MTBF, non-CSIRO usage reaches 40%, CSIRONET: 50 PDP-11 nodes, 48 kbit/s trunk lines, custom host-network interfaces

1975-76 – DCR, building extended, Fernbach report, non-CSIRO usage 50%, CSIRONET: 43 nodes, multi-host development, peripheral driving, NODECODE exported; preparation for 3600 end: Cyber 76 and node enhancements

1976-77 – DCR, Chiefs: Claringbold replaces Lance, 3600 nearly phase out, COMp80 direct connect, typesetting, standalone Cyber 76 system, Ed replaces TED

1977-78 – DCR, 3600/DAD/Document Region decommissioned; 2500 users, Birch report: cost recovery and maximising usage, rationalising Divisional computing, rationalising government-supported scientific computing; improved reliability, CSIRONET: 68 nodes; new computer hall in preparation

1978-79 – DCR, restructure, FACOM joint development project; M-190, NSC Hyperchannel and ATL installed; low-priority charges, improved resiliency, Ed libraries, CSIRONET: distributed services, network links to USA, 96 nodes; new computer hall under construction, visitors to User Assistance

 

1967 – DCR, DAD, Document Region management, disc

  • Division of Computing Research

The newsletter 28 announced on 1 October 1967 that the Computing Research Section would in future be known as the Division of Computing Research – the standard discipline-based unit for CSIRO. But DCR was different – it did do research, but was mainly providing a service.  Dr Godfrey Lance was the inaugural Chief.

C.S.I.R.O. Division of Computing Research – Newsletter No. 28 – 1.10.1967

The interactive system on the CDC 3600 under the DAD operating system allowed interactive users at the consoles to submit short jobs for priority execution, which of course led to contention.  Here is an attempt to put a dynamic time-limit on such jobs, dependent on how long the currently-running job has to go.

This is quite clever – if short jobs are running, then only more short jobs can run, as would be typical during peak periods: at other times, longer jobs are supported.

Another scheduling issue arose with the partitioning of batch jobs into two classes – short and long.  Here’s an appeal to users to help.

The final item in the newsletter appealed to users not to save documents unnecessarily, with the threat of FLUSH always present!
Terry Holden was to present a seminar:

Clearly, there were difficulties in managing such a small space in the face of high demand. 

Charging was an ongoing process – attempts were made to cover costs:

Charging aimed to encourage acceptable behaviour from users, using both carrots and sticks!

The need for charging extended to magnetic tapes, with similar aims.
Note the small size of the disc area available, and the need to take into account blocking, and that discs had gaps, like tape, between blocks of data.

1968 – DCR, widening applications, displays, more branches, evolution of on-line storage and management, first PDP, interactive editing (FRED), teletypes, first remote access, Trevor Pearcey seconded to CDC

It was notable how soon DCR was exploring non-computational uses of the computing facilities – Deciphering Vocal Input was the subject of a report – decades before practical voice recognition.
DCR was also actively learning from other computing centres.
Basic Fortran courses were arranged at major sites.
The use of the display terminals was growing, but there was no remote access.
However, DCR was branching out – the first facility outside a capital city was established at Griffith, and access was provided in Brisbane for the first time.
Finally, modern communications were established with a telex facility!
Peter Hanlon added (private communication Jan 2020): “A later development popular in Queensland provided job submission on Telex tape!”
This is the beginning of the ability of users to retain information on-line on disc drives.  Note the in-built management tool invoked at weekends!  Peter Hanlon (private communication Jan 2020) added:

The disc in question (CDC813) consisted of 2 stacks of 18 platters about 600mm in diameter. 64 of the platter surfaces were used for storage totalling 100 million 6-bit characters. Every weekend it was buffed and polished by CDC engineering staff in readiness for the next week. The disc occupied a box about 1800L X 900W X1700H with a separate box of electronics.

The disc control statements (DRCOPDF etc) were prompted by a source/action/destination mentality, with DFDEL having a particular air of finality.”
This announced the appointment of Mr. J. J. Russell to the position of Director of Technical Services.
It was a wide role, including job scheduling, a perennial problem!
Here’s some more from Newsletter 33.

A new control statement was available under DAD on the 3600, supporting the scheduling of jobs.  It allowed users to specify that the displays had to be inactive for the job to run (perhaps because it needed all available memory), and other restrictions.  For the first time, it allowed job execution to be deferred until after the specified time of day.  These facilities allowing restrictions and deadlines are still important in scheduling today – for example, the NCI job parameter “-lstorage=gdata/<project>”.

Scheduling of interactive work was an early issue – the following shows an attempt to allow interactive work to have long time limits, provided the job did work in smaller bursts.  There was, of course, only one processor available.
The announcement of the acquisition of a PDP-8 in Canberra to act as a controller for teletype terminals marked the start of the CSIRONET network.
Jobstacking was the process of copying card decks to tape at the branches, followed by airfreighting tapes to Canberra. The limitations of this were starting to appear, and advice being offered hinted at things to come in source code management.
DAD continued to be developed, with the ability to manage persistent documents on the disc storage – ‘persistent’ still meant until the weekend!
The management of the location of documents between the drum and the disc was described.  A facility was provided to check if a document was on the drum, and if not, copy it from disc, so providing a primitive automatic recall in a Hierarchical Storage Management (HSM) system.
Notice was given of changes to the plotting routines, used for driving the drum plotters.  However, a poor early decision was noted, with the removal of a Fortran Common block called RHHUDSON, the name of a key staff member!
Peter Hanon added (private communication, Feb 2020):
The reference to rewriting Plot and Text was a major effort and a short-term necessity. The plotters were using disproportionate amounts of storage for spooling, resulting in frequent ‘drum overflows’. RH Hudson and I invented a packing algorithm which resulted in an average 16 times reduction in file size.
One of the later triumphs of Henry Hudson’s career was first announced in this newsletter – an on-line text editor called FRED.
Creative users were continuing to develop utilities and making them available to others to assist in managing their use of the C.S.I.R.O. Computing Network.  Here’s an example from the Division of Animal Genetics, contributing by Brian McHugh who later worked for DCR.  There was not enough on-line storage for storing binaries (executables or apps to the modern generation), so they had to be stored on magnetic tape.  This utility provided random access on a sequential medium.
A demonstration of the use of a  VISTA display terminal was to be repeated – no datashow systems then!
Cards were the principal input medium.  They were stored in the cardboard boxes they came in, or in large metal trays in a special cabinet.  (See the artefacts section for pictures.)The newsletter announced the arrival of equipment for the Perth branch – there was no connection to the East coast, but there was a connection by telephone lines to UWA.
A new utility, DRLIST, was announced: the start of tools for managing data holdings.
The newsletter announced improvements to the login in process to the display terminals, in anticipation of connecting teletypes.
The security was interesting, with the system remembering the previous charge code used!
Here are some pictures of a teletype – an ASR33.
Here’s an example of the use of the displays to show attribute classes.
And, finally, a warning on the use of unlabelled tapes.  This does not refer to a physical labelling of the VSN (volume-serial-number) with a human-readable sticky label, but with the practice of writing onto the beginning of a tape some identification information, particularly the VSN.  This enabled the operating system to check what tape was mounted, and ensure it was assigned to the correct job.  See Wikipedia – Tape label
However, each vendor and operating system had its own ‘standard’ until ANSI Standard X 3.27 arrived.
The 1967-1968 Annual Report can be found here.  The title page contained:
The General section contained the following five paragraphs:
The change of name was seen as a recognition of status.Further branches were coming into operation, with Digital Equipment Corporation (DEC) PDP series being favoured (after the first installation of a PDP-8 in Canberra).
Interesting to note how the word ‘man’ would no longer be used in the context of appointments like the one above and the one below!
Data acquisition (logging) and the following automation was clearly of great benefit to an organisation like CSIRO which at the time was heavily focussed on agricultural work including field data collection.  Small special-purpose computers were fore-shadowed!
And, funds for major expansion were not available.
Under the heading “Central Laboratory Operations” was found:
One of the unanswered questions is when did the Csironet network begin?  Clearly, a network was in mind with Trevor Pearcey and E. A. Cornish’s proposal for the establishment of a C.S.I.R.O. Computing Network – initially the focus was on the four CDC 3000 series machines and keeping them as compatible as possible, with courier and air-freight services being used to collect jobs and return output.  The DAD system allowed interactive usage for the first time, but this was initially confined to the building housing the 3600 in Canberra.  But, the installation of a PDP-8 in April 1968 and the development of an interface to the 3600 enabled teletype access, and the item below shows that one of the teletypes was at another Division (actually not far away on the Black Mountain site).
Peter Heweston (private communication, February 2020) reported being a user of the terminal at Land Research.
A request on the CSIRO Yammer provided the following note:
Other key improvements were highlighted (although it is hard to believe users accepting enthusiastically the actions of FLUSH!):
Under Branch Operations it was noted that:
It is surprising that this was not in place earlier, given that CSIRO Divisions were scattered across the metropolitan areas.
The idea of daily turnaround seems primitive, and yet that is often the case now for users of shared HPC systems that are heavily loaded!
Here’s an example of an application far removed from scientific computation:
The Programming Consulting section had a list of some of the projects DCR staff members had been involved in as consultants:
Peter Hanlon (private communication Feb 2020) added notes on some of the novel applications being run:
    Dr Alan Head from CSIRO Tribophysics (at Melb uni) invented a half-tone photography scheme of crystal diffraction patterns using overprinting of lines on a line printer.
Mr A (Tony) Bomford of the Commonwealth Division of National Mapping made very innovative forays into early mapping of the continent.Various groups from the Bureau of Mineral Resources (John Brown, Roy Whitworth) analysed magnetic anomalies from air and sea surveys.The Commonwealth Department of Works (Hans Kardon) used plotter facilities for cut-and-fill depictions of major roadworks.”
The wheel might have turned – modern science might have more data than it knows what to do with!
The Research Section gave more details of the PDP-8 Teletype System:
This shows that the network was confined initially to the local area, but that a wide-area link was being planned, which would be a true packet-switched network – being from computer to computer, not just a remote terminal.  The description continued, highlighting the design methodologies used to construct the interface, and the use of the 3600 assembler and a simulator.
The editor FRED was described, including its relationship to TECO:
Thousands of people will have used FRED and its successors Ted and Ed, including programming using boxes, which could hold text and/or commands.
Here’s a report on the progress of the speech recognition work:
(Peter Hanlon reported that W. S. Cumpston was a vacation student.)
This newsletter announced a series of seminars on interactive computing, and one by Trevor Pearcey on information handling.
Charge increases were a regular feature in the days of inflation.
Peter Hanlon added (private communication Jan 2020): “The new rates from 1st July 1968 caused consternation amongst the Research staff who managed the accounting processes as a sideline. Until that time there was an integer relationship between CPU seconds and the billing rate, and this departure caused some fundamental change.”
And, a new capability was added of scanning and digitising pictures, and writing the results to magnetic tape.  Also of note is the mention of the interface built by C. S. Wallace: “He was appointed Foundation Chair of Information Science at Monash University in 1968 at the age of 34″.
Several seminars and lectures were announced, in addition to the on-going basic Fortran courses.  The first two were about interactive computing.
The next was about a facility for maintaining images of card decks on magnetic tapes – an early step in source code control systems.The next was an ‘advanced’ lecture on using magnetic tapes and plotting.
Trevor Pearcey’s secondment for a year to CDC in the US was announced – this work led to the STAR computer which became the Cyber 203 and Cyber 205.
The disc maintenance at weekends caused the loss of all stored data.  A procedure was put in place to save the documents to tape at the beginning of the weekend maintenance, and restore them at the end.  Cleverly, the documents were stored on tape in the order of most recent activity, and the restoration occurred in the same order, but stopped short at a threshold.  This presaged the CSIRO IMT Scientific Computing scalable flushing algorithm currently in production use.

DCR went on to develop the storage services, utilising the three tiers of drum, disc and tape, providing a virtual view of the tiers, and providing one of the first Hierarchical Storage Management (HSM) systems, with automatic recall of off-line files.  See the presentation “Pre-history: the origins of HSM in scientific computing in Australia” referenced in the resources section.

The September 1968 Newsletter reported that “Circuitry for the connection of remote Teletypes to the 3600 via the PDP-8 is now complete.  A Teletype using this system is now in daily use at the Division of Land Research in Canberra to execute and edit programs,  The installation of Teletypes at a greater distance awaits P.M.G. type-approval of the equipment and the availability of lines.”

At this point, I’ll mostly leave out items from the Newsletters, and concentrate on the Annual reports.

 

1968-69 – DCR, displays, appeal for expansion, Perth branch

The 1968-1969 Annual Report can be found here.  The cover was:

 

Here are the contents pages:
The first paragraph under General was:
This became a recurring theme over the years: lack of finance for expansion to meet the needs.
Later under General:
This is the further expansion of the network, with remote terminals in locations in Canberra.
Furthermore, the Perth branch became the first to provide a computer to computer link:

1969-70 – DCR, wide-area network=CSIRONET, access to CDC 6600, multi-user disc access, interactive job scheduling, computer-to-computer link, visualisation

contained the following:
so we can confirm a limited wide-area network was in operation in September 1969.  How much this constitutes a network (in the sense of DARPANET) is open to examination.

CSIRONET

DCR developed a packet-switched network from about 1969, the year the DARPANET was started (which became the Internet). Over time, users could have remote card-readers, printers, plotters, and paper-tape readers and punches, and teletypes or visual terminals attached to a local node (PDP-11). Ten characters per second was the standard speed for teletype terminals, with 30 characters per second for display terminals being considered fast. Most CSIRO sites acquired the basic node, card reader and printer, and often had a DCR staff member as operator by the mid-1970s. For example, Kay Challis at Aspendale.

1970

The beginning of 1970 brought an announcement in a CSIRO information circular of access to a CDC 6600 installed in Sydney. The information circular also announced charges being applied to not just computing time and drum storage, but to disc and tape storage, tape mounting and the mounting of non-standard paper on output peripherals.

The 6600 had a peak speed of 3 Mflop/s, compared with 0.35 Mflop/s for the 3600.  (Seymour Cray aimed for each successive system he designed to have an order of magnitude higher performance than its predecessor – part of this was achieved by advances in the hardware, but also by advances in the architecture.  The 3600 was essentially a serial machine (though i/o was offloaded somewhat), the 6600 had multiple functional units allowing simultaneous CPU operations, and the 7600 had pipelined multiple functional units, allowing an assembly-line operation, and had a peak speed of 36 Mflop/s.)

 

The 1969-1970 Annual Report can be found here.  Here are the contents:
Under the heading of “New Services and Equipment”, the 6600 service was highlighted, noting that not much “off-loading” had occurred.
Furthermore, the disc access was revised to allow simultaneous access by multiple programs – something we barely think about today.
Another advance was the provision of a queuing facility for exclusive interactive usage – the previous system would have been frustrating.
The link from DCR to the Division of Entomology was reported.
Visualisation recording was enhanced by an automatic camera attached to a display.
The first CSIRO computer-to-computer link was established in Perth.
The discs on the 3200s enabled system files to be readily-accessible, replacing the former configuration where these files had to be loaded from magnetic tape.
Under the heading of “Research”, the following item gave an insight into the widening field of applications for the CSIRO Computing Network.
Also, the management of the Murray-Darling Basin was clearly under the microscope even at this time.
Finally, under “Research” was an item about echograms (ultrasounds).
And, here is an example of the use of the line printers to provide graphical output – lighter areas are blank, off-white are marked by full-stops, and various other characters are used (together with over-printing) to produce a range of grey-scales.
(Printers were controlled by  ‘carriage-control characters’.  The first character of each line sent to a printer was not printed, but used to control spacing.  A blank resulted in the printer advancing one line before printing, a 0 caused an advance by two lines (double spacing), a 1 caused an advance to the next page, and a + caused no advance before printing, resulting in overprinting.  Frequently, programmers forgot about the control characters, and could waste a lot of paper, which found its way to homes and kindergartens.  On the other hand, people spent a lot of effort digitising pictures and rendering them as ASCII art, e.g. snoopy, the Mona Lisa, Christmas trees.  See ASA Carriage Control Characters )

1970-71 – DCR, two-bank DAD, growth in on-line storage demands – management, Townsville office, job submission via telex, start of network project, interstate lines, PDP-11s, text processing, small computers, DAD exported

 

The 1970-1971 Annual Report can be found here.  Here are the contents:


 

The first item under the “New Services, Software and Equipment” heading was:

It is noteworthy how much of the additional memory was dedicated to system needs, rather than to user programs.  By this stage the CDC 6600 (probably with 64 k of 60-bit word memory, or 492 kbyte) was probably catering for jobs requiring larger memory than the 3600’s 1-bank memory of 197 kbyte.

Once users were able to retain documents on on-line storage, the management of this became a problem, illustrated by changing strategies to deal with it.  It is gratifying to note how the service continued to grow in supporting users, by providing off-line storage for the overflow.

A new office was opened in Townsville during the year, bringing the total to eight.

Publications were moving to digital source, rather than paper, allowing for easier updates to individual pages on paper copies in loose-leaf binders.

The mailing list was computerised, with the staff member using the on-line text editor FRED.

Townsville was like Perth, in having a PDP-8 to provide access to a nearby university’s PDP-10.

Who would think now that people would use a service to submit jobs using paper tape and a machine on the telex network, which was primarily used for business communications!

The following item gives an insight into the difficulties of maintaining software for remote un-connected machines.

The big item for the year was the commencement of the network project, based on earlier work involving the PDP-8 front-end to the 3600.  This planned to allow remote job entry and output to the 3600 in Canberra from the branches, and allow remote interactive access for the first time.

CSIRO was one of the first organisations to make use of the interstate telephone network for data traffic – a speed of 2400 bit/s to each node was planned.

The next item showed the pioneering emphasis on speech processing
– Hello CSIRO perhaps, rather than Hello Siri or Alexa or Google.

CSIRO wrote its own monitor (operating system) for the PDP-9 at Griffith, to allow time-sharing.

DCR built its own spell checker for its documentation – it built a lexicon from existing technical text, and used that to check any new or revised documentation.   Hash-table techniques were used, even at this early stage, to allow proof-reading to execute in 1 – 2 seconds per page.

DCR was facilitating the use of small computers in CSIRO Divisions, and hosted a 3-day workshop in November 1970.  This led to study groups to work in specific areas.  Doubtless, some of the work found its way into the future CSIRONET network facilities, and some led to heavy usage in Divisions of local computers for data acquisition, etc.

As well as the porting of DAD to the system at Computer Center Santa Barbara as reported in chapter 2, DAD was made available to CDC, and was implemented on a CDC 3800 in the USA with the assistance of Peter Ewens.

1971-72 – DCR, CSIRONET in operation, PDP-11 nodes at branches, Rockhampton branch, virtual Document Region, reentrant FRED, improved reliability, Claringbold acting chief

 

The 1971-1972 Annual Report can be found here.  The contents page is similar to previous years, and there were no new branches (but see below):The network came into operation under the name CSIRONET: Here is the list of consoles: most major CSIRO sites had access, along with several Government Departments.

 

Charging for access to the CDC 6600 was revised downwards, but would, I guess, still be a disincentive to usage.

Sorting is one of the fundamental algorithms for computing:  DCR provided improved routines with vast improvements over naive algorithms.

Despite the listing of branches under the contents, there was a new branch started in Rockhampton.

The major branches acquired PDP-11s to act as nodes.  The start of service date of April 1971 seems at odds with the CoResedarch announcement of service commencement in April 1972.

CSIRO wrote its own monitor (operating system) for the PDP-8/I in Perth, to allow time-sharing.

The summary of fields of research gives some idea of the breadth of the applications of computers at the time. Development of DAD continued: additional disc and channels, support for up to 70 interactive ports, improved scheduling, reentrant code, and the virtualisation of the Document Region spanning drums and disc, with overflow to magnetic tape,; and system reliability (to support interactive use).


Finally, the Chief, Dr Godfrey Lance took up a position in London, and Dr Peter Claringbold became Acting Chief – a sign of things to come later in the 1970s.

CoResearch 157 April 1972 announced the start of operations of SIRONET (later CSIRONET, then Csironet).

Interactive access was supported by a CSIRO-developed text editor called FRED (then TED and later ED). Henry Hudson was the main developer. Versions for the Cyber 76 (Ed, using the additional LCM) and FRED for PDP-11s were developed).

 

1972-73 – DCR, Cyber 76 announced and arrives, Margaret Thatcher visits, DR retention periods and becomes and HSM, extension of CSIRONET to provide local nodes, TED replaces FRED – boxes

CoResearch 162 September 1972 reported on a visit by a UK Minister who became more famous.

The big news from 1972-1973 was the announcement and arrival of a new CDC Cyber 70 model 76 (Cyber 76).

(Two 7600s.  3D rendering with a figure as scale: from reference CDC 7600)

CoResearch 163 October 1972 announced a “New Computer”:

CSIRO acquired the best supercomputer of its time in 1973. The Cyber 76 was a re-branding of Seymour Cray’s 7600, itself a breakthrough machine with multiple parallel pipe-lined functional units, and a follow-on from the CDC 6600. The Cyber 76 had a clock rate of 27.5 ns, had 64 k of 60-bit words main memory (SCM – small-core-memory), and 256 k of 60-bit large-core memory. This was later extended to 512 k words, to support interactive usage, and could comfortably support 100 users across the country doing editing, etc.  Its peak speed was 36 Mflop/s.  See the reference CDC 7600.

CSIRO acquired the system with Federal Government money, and was required to continue to give access to other government departments. The system was used for nearly every computing task in CSIRO – payroll, libraries and all.

If was rumoured that the CDC salesman received the cheque for the purchase (~$5M) and caused a stir when he banked it into his account at an ordinary suburban bank branch in Canberra. The next day, he had to forward most of the money to the company.

CoResearch 170, July 1973, highlighted the arrival.

The September 1973 Newsletter noted that:

Cyber 76 installation started 9 June
Acceptance tests started 29 June
Acceptance tests completed 13 July
CSIRONET/Cyber magnetic tape link operational 18 July
CY link operational 13 August (DAD system – SCOPE 2.0 system)
CO link operational 22 August (SCOPE 2.0 system – DAD system).

(SCOPE 2.0 was the operating system on the Cyber 76).

Also noted:

“The communications lines of CSIRONET are currently:”

– 8 links were listed between the mainland capitals and Cronulla, Rockhampton and Townsville, with speeds of 2400 bps or 4800 bps (bits per second).

 

C.S.I.R.O. Division of Computing Research – Newsletter No. 92 – May 1973 contained a lot of information about the upgraded Document Region.  Retention periods were introduced for documents, allowing users to specify how many days each document was to be retained.  But, there was a sting in the tail: charges for the full retention period were imposed at the time a retention period was specified!  This encouraged the use of short retention periods, though these could be extended (with immediate charges for the extended period!).

Copies of unexpired documents were made onto tape weekly.  If the document’s retention period was long enough, it was copied to two tapes (“for security reasons”).  After this, the document may be flushed.  These are exactly the HSM processes which are in operation on the CSIRO Scientific Computing Data Store today – migrate data from disc to tape, and free up disc space for migrated files – except that retention periods are not defined, and are taken to be infinite.

And, here is the counterpart to migration – automatic recall or retrieval. And, there was integrtion with the job scheduler, and batching of requests for reasons of efficiency.

Finally, users were able to view their holdings in the system, and issue explicit requests for recall or retrieval – the equivalent of the dmget command which has been in operation on CSIRO SC systems from 1991 to today.

And the previous explicit use of the disc storage system was being phased out to make room for the integrated automatic document region.

The following manual was an attempt to make the main manual more accessible!

Charging was a recurring theme: this item described revisions which were made to make the charges more realistic rather than relying on sampling.


The 1972-73 DCR Annual report can be found here.  Here is the contents page, and the list of branches – now including Rockhampton and Armidale, bringing the total to 10.

There was more information on the Cyber 76, with some details of the major changes to infrastructure to support it.

As well as funding for the Cyber 76, funding was in place to expand CSIRONET, so that most Divisions would have remote job entry and interactive access.

 

Development continued on the DAD operating system, primarily to support the Cyber 76 and the growing network

The Document System continued to be enhanced, supporting user-specified retention periods, and automatic retrieval from tape, thus providing one of the earliest Hierarchical Storage Management system.

The FRED editor which was written for the early display consoles, was replaced by TED, which supported  a range of terminals, and had extended features.


The last item mentions Ted boxes.  These were storage areas within the editor which could store strings of text, useful for copying and pasting text (before copy and paste were invented).  However, a key feature was that as commands to the editor were strings of text, these commands could be stored in boxes, and executed like a macro: but these supported one box calling another, and recursion, and many other features.  Large software projects sprang up using the features of TED boxes, and one group in Admin. even produced an Ed Programming Language on top of the Editor.  Utilities such as permanent file backup on the Cyber 76 were written in Ed boxes.  This could be accessed with commands like:

U/PUBLIC/[/PFBACKUP/$

There were Ed (the successor to Ted on the Cyber 76) programming contests!

Some primitive level of security was introduced.

Here’s an insight into the challenges of writing software for the PDP-11s that were used as nodes on the network – all in assembler, and cross-compiled from the Cyber 76.

 

1973-74 – DCR, Cyber 76 in service: links, porting apps, permanent file management; CSIRONET: 30 nodes, delays in lines for network, Remote Access Group; AI, 3200s removed, building extension announced

The 1973-74 DCR Annual report can be found here.  Here is the contents page (the number of branches remained unchanged).

The introduction more than hinted at the difficulties faced by the Division in bringing the Cyber 76 into service: chief among these were the work faced by users in converting to the new machine.  (In a conversation in April 2020, Peter Heweston mentioned that each CDC machine came with a new operating system that required users to convert what we now call apps.)  At that time, IBM was settling on one OS, and in later years we have the near duopoloy of UNIX/Linux or Windows.)  Further problems arose from delays in the provision by the Post Office of transmission lines, the problems of maintenance of the PDP-11s being used for intensive communications tasks, delays to new accommodation and staff shortages.  On a positive note, the year was the first of a three-year expansion program, and Regional Computing Committees were set up.  Finally, the introduction included the goal “to provide the Australian scientists a computing service comparable to the best in the world.” 

 

Under the heading “New Equipment, Services and Software” was an item on Operational Development of Cyber 76.  This describes the transition from a magnetic-tape regime to channel connection to and from the 3600, which acted as a front-end.  It also describes pressure on the permanent file system, almost from the start.

 

 

Also described was the expansion of CSIRONET.

 

Documentation took advantage of the new technologies – there was no what-you-see-is-what-you-get visual access though.


Interactivity with the computers was enhanced with the ability to run jobs on the Cyber 76 directly from the TED editor (running on the 3600), with output returned to the editor.  The charging rate was about triple the rate for batch processing!

The growing importance of the network led to a separate group being formed.
Several projects were described under the Research section.  This one, about bushflies, harks back to CSIRO’s work on developing insect repellant after flies annoyed the Queen on a visit to Australia.

Database systems were in their infancy, but DCR was heavily involved in providing access to such systems, albeit in a manner that would be considered unusual these days – through Fortran!

And, given the current emphasis on machine learning and artificial intelligence, it is an indication of the pioneering thinking that DCR co-sponsored a 5-day symposium on AI in 1974!

Under the “Summary of Computer Equipment” section, the definition of a basic node of the CSIRONET network was noted.

The report noted that ‘The 3200s were removed by October 1973.”  They were in service for about 9 years.  There was no other mention in the annual reports of their demise, although the previous year’s report mentioned that the 3200s had been connected to CSIRONET.

Finally, the front cover showed the front of an unnamed equipment rack.  The back cover showed the rear of the cabinet, showing the huge bundles of wires.

 

1974-75 – DCR, COMp80 in service: microfiche, image processing, Hobart branch, Cyber 76 low MTBF, non-CSIRO usage reaches 40%, CSIRONET: 50 PDP-11 nodes, 48 kbit/s trunk lines, custom host-network interfaces

The 1974-75 DCR Annual report can be found here.    The front cover featured a satellite image

with the legend:

A prologue set out DCR’s charter, and had a curious reference to ‘a user population with varying skills, varied attitudes and requirements’.  I had joined CSIRO at this stage, so might have been covered by this statement!

Here are the contents. (The number of branches increased by one with the addition of Hobart, bringing the total to 11.)

There was again a long introduction, but this year’s one was attributed to G. N. Lance Chief of Division.  Progress was being made on the building extension.

The Cyber 76 was at least more reliable than the CDC 6600, which reportedly had a MTBF of 17 hours!  With Seymour Cray’s design goal for each machine to be ten times faster than its predecessor, reliability was not the main goal.  And, the SCOPE 2 OS was a special for the 7600/Cyber 76 – the first 7600s were delivered with no operating system, which was the responsibility of the customer.  See CDC 7600 at wikipedia: “Both LLNL and NCAR reported that the machine would break down at least once a day, and often four or five times.”

 

The coming of the COMp80 brought a revolution in the ability to store output and produce large volumes of graphics.

CSIRONET (the network) continued its evolution, with 48 kbit/s inter-capital links, and all sizable user groups having an on-site or nearby node.

 

DAD and the 3600 continued to support interactive and central input/output facilities.

From its inception as the C.S.I.R.O. Computing network, access was to be provided to government departments, and it was noted that non-CSIRO work had reached 40% of the total.  This may be the start of the difficulties CSIRO had within its charter of reconciling its research and application roles with providing a service to others which was not part of its mandate.

The introduction finished with an expression of hope that the Cyber 76 will become a computer on which reliable interactive computing is feasible, and this was to unfold within a couple of years.

Under the New Equipment section, more details of the COMp80 were given.  (A later section provides more information on the use of microfiche.)

Here is an example of DCR staff needing to modify SCOPE 2 to support expanded storage.  This was also the beginning of having additional on-line storage under user or group management – no backup, no restoration in the event of loss, no flushing, but putting the responsibility back onto a group of users.

 

The business of managing data on unreliable media with unreliable reading and writing devices is so far removed from the storage services available to today’s generation.  This item notes improvements with newer equipment.

CSIRONET continued to expand in size, in speed, in resiliency and reliability.


DCR continued to embark on targetted hardware development to support its unique network.  The last paragraph hints at future developments to replace the 3600’s roles.

 

Managing shared storage areas continued to require attention, this time on the Cyber 76

 

By 1975, Csironet was well established.  A paper by Colin D. Gilbert, The Evolution of Csironet, noted that:

The paper included a map showing the extent of coverage and the connection path from the Cyber 76 to a sample node:

and a list of nodes:

The January/February 1975 Newsletter announced that a “COM Unit is Ordered”.  The installation was announced in July 1975. The COMp 80 provided output to microfilm, microfiche, 35 mm film and bromides.  Microfiche became the default way to store large amounts of output for Csironet users.

Here is a picture of some microfiche, with a ruler beside to show the scale:

If you zoom in on the picture on the left hand side, you could see that each fiche contains images of hundreds of pages – the capacities were 420 A4/quarto size, or 280 11″x15″ line printer page images.  The upper fiche is an original, and the lower one a duplicate, which cost only a few cents more.  These were read with a reader/viewer, like this one:

Most staff heavily involved in computing had ready access to a viewer, and many had their own on their desk.

Microfiche was a revolution in enabling large quantities of output, including text and graphics, to be stored for reference.  Microfiche was ideal for distributing user documentation economically: for the first time, each user could have access to all the documentation (which at this stage was far too large to be stored on-line!).  Here is a picture of microfiche making up most of the Csironet documentation from 1987, stored in a special holder in a loose-leaf folder.  If you cared about the originals, you would use the cotton gloves pictured, to preserve the contents if they were to be subsequently duplicated.

 

 

 

There were plans to install two CDC Cyber 172s to take over the front-end role of the 3600 (DCR Newsletter 104, June 1974), but these plans were abandoned.  Instead, the 3600 continued in service for several more years, extra LCM was installed on the Cyber 76, and an interactive subsystem (CYI) was built to run on the Cyber 76, along with a new version (Ed) of the TED editor.

 

1975-76 – DCR, building extended, Fernbach report, non-CSIRO usage 50%, CSIRONET: 43 nodes, multi-host development, peripheral driving, NODECODE exported; preparation for 3600 end: Cyber 76 and node enhancements

The 1975-76 DCR Annual report can be found here (78 pages).    The front cover showed a view of the building with the newly-completed office wing, and a view of the computer room.

 

 

The foreground shows rows of CDC model 844 disc drives which featured removal disc packs.   There were 31 units, and each disc pack had a storage capacity of 118 million characters (88.5 Mbyte).  The background shows magnetic tape drives.

The Division’s charter remained unchanged since its inception, and was re-stated:

There followed a list of applied research areas undertaken by the Division.

 

 

The next paragraphs went on to describe how the service addressed the needs of scientists.

 


The above covered the activities of the Division, but gave a glimpse of the need to continue to emphasise the research as opposed to the purely service role, which was the majority of its activities, but was not a role that was core to CSIRO’s charter.  As I saw it, this tension remained to the end of the Division, when it was split into the service arm Csironet and the Division of Information Technology.

The Chief’s report then followed, with a fairly pointed criticism of the framework under which the Division was constrained.

The Fernbach report was one of many over the years.

The following observation highlighted a milestone in the growing proportion of the computing done by non-CSIRO users: unless this was considered part of CSIRO’s mandate to help industry, it was not CSIRO’s core business, and in any case, most of the non-CSIRO usage was from government departments.

More storage was acquired, and Csironet continued to grow.  The CDC 3600 was clearly nearing the end of its life, but was a core component of the network and services.  Many of the subsequent pages of this annual report highlighted work underway to replace this functionality with services run primarily on the Cyber 76 under SCOPE 2, which was never designed for such interactive, network-focussed, real-time, and direct user interaction.  It was a challenge.

Some of the new facilities under development included:

  • a simulator for the Cyber 76 SCOPE operating system, allowing the withdrawal of the 16:30 test period
  • development of a Cyber 76 Peripheral Processor programs to interface with CSIRONET – batch, interactive and test
  • development of a ghost station on the Cyber 76, to replace the 3600
  • integration of local modifications (23,000 lines) into each new release of SCOPE
  • development of Cyber 76 replacements for TED (interactive) and RIO (remote input/output).  These were developed as user-level jobs, with additional locally-implemented privileges
  • the creation of an event processor, which allows messages to be passed between jobs
  • development of the Cyber 76 hardware Maintenance Control Unit to allow it to be used for software maintenance and as an operator console.
  • automated pack assignment
  • automated tape label writing
  • magnetic tape reservation
  • magnetic tape audit and inventory
  • accounting changes, including showing itemised resource usage and cost for each job – see the example Cyber 76 job dayfile listing at here.
  • generalisation of the accounting systems
  • interactive preprocessors, particularly for Document Region (3600) and permanent file (Cyber 76) backup
  • generalisation of CSIRONET, to support multiple hosts (not just the 3600), and to provide connection to the Cyber 76 as one of the hosts
  • connection of the COMp80 to the central hosts
  • System Performance Analysis and Resource Monitoring
  • Reference Indexing Interrogation

The last item (by Peter Hanlon) involved the development of an indexing tool to collect keywords from publications and store them and their location in a database.  For example, the Users’ Guide of some four hundred pages yielded an interrogation file of approximately fifty thousand words.  Querying this typically took 20 milliseconds of CPU time.  This is a local version of Google search, 20 years before its first search capability in 1996!

 

And, here is another export success:

 

Here’s another major piece of work allowing the network to become multi-host:

Research work was reported.  Two items of particular interest were the modelling of social welfare policies, and modelling of carbon flow in an estuary to understand the ecosystem.

 

Terry Holden reported on the Review of Computing Services by Dr S. Fernbach.  The summary of the report’s findings were included. (The full report is in the Terry Holden papers).  Here is the summary.

This addresses some of the users’ concerns – “failure of DCR to provide reliable, stable and adequate computing support”, and lack of “hardware and software” desired by users.  Some of the solutions involved additional funding and a 50% increase in staff, which never eventuated (see the graph of DCR staff numbers, which shows a decline in staff numbers for each of the next three years!)

 

1976-77 – DCR, Chiefs: Claringbold replaces Lance, 3600 nearly phase out, COMp80 direct connect, typesetting, standalone Cyber 76 system, Ed replaces TED

The 1976-77 DCR Annual report can be found here (89 pages).   The contents were similar to previous years, except the report started with “Report of Past Chief”, given below in full.

 

 

 

 

 

 

 

 

 

 

 

Thus in 1977, Godfrey Lance handed over the Chief’s role to Dr Peter Claringbold, following the failure to gain more resources from the CSIRO Executive to meet the growing service load.  To what extent he was forced to do this may remain unknown.  He remained as Assistant Chief in charge of research, but departed soon after.  He was at the Avon Universities Computer Centre, Bristol, UK in October 1980 (csiro dms Newsletter no. 70, November 1980), and lived in Bermagui NSW in retirement.  He died in 2020 – see obituary.

This was a watershed time on the history of the Division.  Godfrey Lance went public through newsletters and annual reports with his frustration with the CSIRO Executive, which would not have been well received.

General access to the CDC 3600 was disabled in May 1977, with most of its functions by then transferred to the Cyber 76.

Apart from the Chief change, the major work reported was headed by:

 

With the COMp80 now in production and on-line to the Cyber 76, typesetting became a major new service.

Picture processing was enhanced with the relocation and expansion of a PDP-11/40.  To help with software development, Don Fraser wrote a Fortran version of the DCR Ed editor – I have a copy!

...

C  THIS PACKAGE IS WRITTEN IN PDP11 FORTRAN TO PROVIDE, IN A SHORT TIME,TED   60

C  A MINI TEXT EDITOR WITH GENERALLY THE SAME COMMAND STRUCTURE AS 'ED',TED   61

C  HENRY HUDSON'S EDITOR FOR THE CSIRO CYBER 76 COMPUTER.               TED   62


Here’s an example research project using the picture processing capabilities.

Here’s another interesting application!

The Summary of computing equipment reported 392 connected terminals, 6 dial-up ports, and 18 auxiliary computer connections.  It listed the nodes and the equipment at the DCR-managed ones.  One of the nodes was at the Reserve Bank in Sydney.

1977-78 – DCR, 3600/DAD/Document Region decommissioned; 2500 users, Birch report: cost recovery and maximising usage, rationalising Divisional computing, rationalising government-supported scientific computing; improved reliability, CSIRONET: 68 nodes; new computer hall in preparation

The CDC 3600 system was de-commissioned in the second half of 1977. One of the features was the DAD operating system, written by CSIRO and Control Data staff members.  It included a Document Region – this was a storage area with multiple levels and automatic transfer of data between the levels – the drum, disc, and operator-mounted tapes (7-track), and so was an early Hierarchical Storage Management system from 1973, with automatic recall of off-line files and integration into the job scheduling.

The 1977-78 DCR Annual report can be found here (80 pages – original was 76pp + covers, but some pages are repeated and pairs of landscape pages are rendered as single pages. Pages 1-34, then 43-44, then 33-76 are present).

The introduction highlighted DCR’s mission.  It continued to have a mandate to meet the scientific computing needs of institutions funded by Government.  This foreshadowed in some respects the role of APAC and NCI, though their remit included more specifically the university sector.  A total of 2500 users was a significant community.

There was an expanded contents section, with major headings for the research areas.

The Chief’s review commenced with notes on new policies, which came from the recommendations of the Birch report.

Recommendation 80 affirmed cost recovery, but added the rider of seeking to maximise usage.  This led to changes, reported below, to open up the systems outside business and attended hours, and striking a much lower charging rate for this.

Recommendation 81 attempted to constrain the acquisition of local computers, and was doomed to failure.  I remember Art Raiche telling me that when his Division acquired their own minicomputer, they could afford to make mistakes and could do science again.  The attraction of owning a bright shining object, and having it available for instant gratification without the strictures of batch systems and queuing and under local management, was very powerful, and still is!

Recommendation 82 attempted to bring centralisation to academic scientific computing, and was again doomed. It was another 20 years or so before the universities agreed to a consortium, driven by government funding to APAC and NCI.

Recommendation 83 tried to confine DCR’s research efforts to areas that users desired.  However, with 2500 users, that was also doomed.  It wasn’t until the mid-1980s that DCR was split into the Division of Information Technology and Csironet for the separate roles of research and services.

 

It was frustrating as a user in those years to have the experience (it seemed frequently) of losing the current work when the network or host system failed.  DCR worked on providing better reliability, and the Cyber 76 version of Ed could recover an editing session if the Cyber 76 or network went down (planned or unplanned).

The above addressed the issue of costs, asserting that CSIRONET’s charges were lower than comparable services, and they probably were.  But, scientists crave freedom of access to facilities, and are reluctant to spend on things that they can acquire freely.  So, for example, some scientists in this era chose to go to other institutions, for example NCAR in Boulder Colorado, since access to the computing services was free there.

Here are some of DCR’s actions to help with the aim of maximising usage (while still retaining revenue.  Many Divisions and user made heavy use of the non-prime-time regime with its considerably lower charging rate.

Here are the job request lines from a couple of Cyber 76 jobs from this era – in the early 1980s, lower case was supported:

r50b(P0017,T20000,ms120000,dab00)        <<<<<     Robert Bell     >>>>>
RUNG21E(P0111,T20,DAB01)                 <<<<<     ROBERT BELL     >>>>>

The first job was a main run, requesting 20000 (octal!) seconds, and a priority of 0017 – again octal, and taking advantage of the priority zero rate.  The second job was a post-processing job, with a higher (but still low) priority.  Both jobs could be submitted together, and because of the dependency specification, the second job would not start until the first job had signalled that its data was ready with a statement like transf(rung21E).

The passing of the CDC 3600


Both the Communications Systems Group and the Csironet network continued to expand, with staff now numbering 13, 68 nodes in operation, 584 connected terminals and a growing number of PDP-11s in service.

Support for card punches and cassette tape drives on the network were new items, illustrating the passing of the old and the introduction of consumer storage devices, allowing a maximum of 263000 characters per side.

A User Assistance Section was formed, led by Barry McDowell, with one other staff member.

The Documentation Section was expanded from 4 to 6 staff.  Microfiche was being used extensively to allow updated documentation to be produced and distributed cheaply to users.  There were three series of publications: Computing Notes, Services Notes, and DCR Manuals.  The newsletter was changed from monthly to bi-monthly, the weekly Cybarite machine stored news sheet and the irregular Cyberflash continued, along with an introductory kit.

The SCOPE 2 operating system on the Cyber 76 supported permanent files – meaning files that could be requested to remain after the completion of a job.  The term ‘permanent’ was unfortunate, because crashes could lose them.  There was a story that one of the operators restarted the Cyber 76 with an incorrect date set, and happily answered the prompts to remove thousands of files marked as expired.  The permanent file facility did not allow for many files, and this became a problem.  Flushing was in operation, and a metric named ‘moribundity’ was used to select the most dormant files for removal.

With local nodes in operation, two DCR offices were closed, with operations being taken over by the local Divisions.

Finally, another expansion of the buildings was announced – 401C.

1978-79 – DCR, restructure, FACOM joint development project; M-190, NSC Hyperchannel and ATL installed; low-priority charges, improved resiliency, Ed libraries, CSIRONET: distributed services, network links to USA, 96 nodes; new computer hall under construction, visitors to User Assistance

The 1978-79 DCR Annual report can be found here (82 pages).

========================== under development below

 

Finally, I’ve left out any of the notes on the applications, which continued to increase in diversity – descriptions were given on projects involving studies of potatoes, of arid land, of wool crimp, of mapping forest fires, and of the structure of bacteria.

1980

By 1980, many CSIRO Divisions had their own smaller computers.  The CSIRO Division of Mathematics and Statistics November 1980 Newsletter describes the start of local computing in 1974 using DEC PDP11/10s and HP 9830 programmable desktops, through to the adoption of the UNIX timesharing system in 1980, using a version developed at the Universities of New South Wales and Sydney (the Australian UNIX Share Access Method (AUSAM)).  Features of UNIX were described, and another article  addressed “Word processing in UNIX”.

1981

1982

1983

Here’s a picture of the DCR/Csironet machine room in 1983 (from NAA), showing the Cyber 76 in the background and many disc drives.  These are probably CDC 844s, which held up to 118 million 6-bit characters on a removal disc pack.

The operator is sitting at a Cyber console with a printer, with a stack of magnetic tapes.  The next picture, taken from beside the operator, shows some of the magnetic tape drives.

The tape drives had a window that slid down from the top, allowing operator access to remove a tape and/or load a new tape on the right-hand hub.  When done, the window closed, and the plastic ‘safety-belt’ around the tape was blown off – there were air-holes in the plastic – allowing the end of the tape to be found, which was then automatically threaded down into and up the right-hand vacuum loop, then across the head in the upper centre, then down into the left hand vacuum loop and up, and then wound onto the take-up spool.  (In earlier versions, operators had to do all the threading manually, and the tapes were kept in hard transparent two-part cases.)  The vacuum loops allow the tape drive to stop and start the tape very suddenly, without breaking it – the loops of tape in the vacuum areas cushioned the tension.  Here’s a picture of the tape library, prior to automation.Here’s a view of the disc drives, showing the discpack cases.

The central row would have stored less than 1 Gbyte!

  • Fujitsu M series

From the early 1980s, CSIRO acquired a series of Fujitsu (FACOM) M series systems – IBM mainframe compatible systems. These were not widely used by CSIRO science staff, but were used for administration processing, and by government departments.  Peter Hanlon (private communication Feb 2020) wrote: “The Facom stuff was actually a golden era, and deserves a lot more press.”

  • Terabit File Store, Calcomp/Braegan Automatic Tape Library

CSIRONET developed a Terabit File Store, using a Calcomp/Braegan Automatic Tape Library (using 6250 bpi 9-track round tapes). The file store provided a filesystem interface to various hosts, with a FACOM system acting as the host. The TFS did not provide HSM – access was solely by explicit command.

  • CDC lower Cyber series

CSIRONET acquired a series of lower Cybers to act as front-ends – Cyber 700 and 800 series. These ran the NOS operating system, which was not popular with SCOPE 2 users (Cyber 76),  For example, filenames were 1-7 characters, not the 1-40 characters supported by SCOPE 2. One Control Data support staff member said the good thing about SCOPE 2 was that CDC designed it and then left it alone (from Dave Micklethwaite). NOS had messy filesystem access (direct and indirect files), and all the early Cybers had to have 8-bit ASCII character handling grafted on.

1984

  • CDC Cyber 205

This was an evolution of the STAR (String Array Processor) and Cyber 203, and was installed in Canberra in mid-1984. It was a vector processor, but memory-to-memory, not register-to-register like the Cray vector processor. It had an elegant architecture, and a very rich instruction set. It used 64-bit descriptors to represent vectors in memory – 16 bits for the length (in 64-bit words, so a maximum of 65535), and 48 bits for a starting address, addressed down to the bit level. The vector instruction set included instructions like add two vectors and store the result in a third vector. There were 8 ‘G’ bits in the instructions that provided extra control: for example, whether to do 32-bit or 64-bit operations, whether to mask the results storage with a control vector, whether the operands are in fact scalars to be propagated, whether to do sign control (e.g. to do the square root of the inputs, or the absolute value of the inputs). A ‘feature’ was the ability of the Fortran language to allow direct calling of the hardware instructions with a Q8 call. The machine at CSIRONET had a peak speed of 400 Mflop/s.  It served until 1990.

The Cyber 205 had a novel memory-to-memory vector architecture.  Programmers could call vector instructions directly.  Here’s some sample code, implementing

C(1:N) = ADUM * A(1:N) + B(1:N) [before that Fortran 90 syntax was available].

DESCRIPTOR DA, DB, DC

Assign da, a(1;n)

Assign db, b(1;n)

Assign dc, c(1;n)

alocal = adum

CALL Q8LINKV (X'08')

CALL Q8MPYSV (X'90',, alocal,, DA,, DC)

CALL Q8ADDNV (X'80',, DB,, DC,, DC)

A footnote:  In the late 1960s, Trevor Pearcey was seconded to Control Data Corporation in the USA, and worked on the hardware design of a new computer called the STAR – String Array Processor.  This became eventually the CDC Cyber 205.

1985

The Cyber 76 was de-commissioned in 1985 after the Cyber 205 went into service.

CSIRO and Csironet were keen to widen the use of the 205: a grants scheme was set up to allow university users to gain access. See Csironet News no. 187: “A User’s Experience of the Cyber 205” by Bob Gingold.

The DCR Newsletter and its successor CSIRONET News contained a wide range of articles, and was brave enough to allow letters and comments critical of the services and policy to be published.  There were anonymous columnists – later unveiled as Peter Hewston, Rob. Bell and Charles Johnson.  Some of the big issues were the rise of mini, micro and personal computers, the charging regime, the language wars (Pascal, etc), source code management (the UPDATE utility), the unreliability of the hosts and network, job scheduling and funding.

After the arrival of the Cyber 76, a weekly on-line newsletter called CYBARITES was provided.

Users could also obtain access to the minutes of the DCR Design and Development (DAD) Committee meetings.

 

Here is a link to a video on Csironet from 1985.