Differences between revisions 6 and 7
Deletions are marked like this. Additions are marked like this.
Line 17: Line 17:
== Starlink Site openings ==

|| Opened || Site || Computer ||
|| 1979 Dec || RAL || 780 ||
|| 1980 Jan || ROE, RGO || 780, 780 (RGO transferred to CAM 90/4) ||
|| Feb || Cambridge, UCL || 780, 780 ||
|| Jul || Manchester || 780 ||
|| 1982 Mar || Durham || 750 ||
|| Oct || Birmingham || 750 ||
|| 1984 Nov || Leicester || 750 ||
|| 1985 Mar || Jodrell Bank || 780 at MAN moved to JOD, new 750 at MAN ||
|| 1986 Apr || St Andrews || microVAX ||
|| May || Preston || microVAX ||
|| Sep || Southampton || microVAX ||
|| 1987 Feb || Cardiff || microVAX ||
|| Jun || QMC || microVAX ||
|| Aug || Keele || microVAX ||
|| Nov || Oxford || microVAX ||
|| Dec || Armagh, Belfast || microVAX ||
|| 1988 Feb || Database (RAL) || microVAX ||
|| 1989 Oct || ICSTM || microVAX (first DECstation at RAL) ||
|| 1990 Mar || RAL || split into 2 clusters: Project (PRO) and Astrophysics (RAL) ||
|| Apr || RGO || closed (moved to CAM) ||
|| May || Sussex || microVAX ||
|| Sep || Kent || microVAX ||
|| Oct || Hatfield || microVAX ||
|| 1993 Aug || Liverpool || ||
|| Dec || Glasgow || ||
|| 1994 May || MRAO || now regarded as a separate User centre from CAM ||
|| Oct || Sheffield || ||
|| 1995 Oct || Bristol || ||
|| 1997 Oct || Leeds ? || ||

== The move to Unix ==

|| Site || Planned || Slipped to || Achieved ||
|| ARM || 94 Jan || 94 Dec || 94 Dec ||
|| BEL || 94 Jan || 94 Dec || 94 Dec ||
|| BIR || 94 Dec || || 94 Dec ||
|| CAM || 94 Apr || || 94 Apr ||
|| CAR || 93 Oct || || 93 Oct ||
|| DUR || 94 Dec || || 94 Dec ||
|| EDI || 94 Dec || 95 Jan || ? ||
|| GLA || 93 Nov || || 93 Dec ||
|| HAT || 94 Dec || || 94 Dec ||
|| IMP || 94 Dec || || 94 Dec ||
|| JOD || 93 Jun || || 93 Jun ||
|| KEE || 92 Dec || 93 Jun || 94 Mar ||
|| KEN || 94 Dec || 95 Jan || 95 Feb ||
|| LEI || 94 Dec || || 94 Dec ||
|| LJM || 93 Sep || || 93 Aug ||
|| MAN || 94 Jan || 94 Dec || 94 Dec ||
|| OXF || 94 Oct || 94 Dec || 94 Dec ||
|| PRE || 94 Dec || || 94 Dec ||
|| QMW || 94 Dec || 95 Jan || ? ||
|| PRO || ? || || 94 Dec ||
|| RAL || 93 Oct || 94 Mar || 94 Mar ||
|| SHE || 94 Oct || || 94 Oct ||
|| STA || 94 Dec || || 94 Dec ||
|| SOU || 94 Apr || 94 Jul || 94 Jul ||
|| SUS || 95 Apr || || ? ||
|| UCL || 94 Dec || || 94 Dec ||



== Starlink Milestones ==

|| 1978 || Panel on Astronomical Image and Data Processing (Disney) set up by Astronomy II committee of SRC. Its report (Jul 79) led to Starlink. ||
|| 1979 || SRC approved setting up of Starlink. ||
|| || First meeting of SAG (Scientific Advisory Group) (Sep 12). ||
|| 1980 || User service began (Apr 1). ||
|| || Official opening (Oct 24). ||
|| 1982 || Tiger Team (Culhane) set up to seek more manpower from ASR Board. Unsuccessful. ||
|| 1983 || Computing Review Working Party (Culhane) set up to decide how to implement cuts in Starlink staff & finance. ||
|| 1984 || Central management moved from Computing Division to Space & Astrophysics Division at RAL. ||
|| || SUC (Starlink Users' Committee) replaced SAG. ||
|| 1986 || Starlink Project recommended to community that IRAF be adopted as Starlink's environment. ||
|| || Software Environment Review meetings. Decided to develop ADAM. ||
|| || Starlink Questionnaire No 1 (Wallace) (SGP/35). ||
|| 1989 || Tiger Team interviewed Starlink staff at RAL to set priorities for Project's activities. This was part of SUC's "Making Ends Meet" study. ||
|| 1990 || ADAM support Group set up at RAL (Allan). ||
|| 1992 || Starlink Review Panel (Willmore). Reported Sep 92. ||
|| 1993 || National Audit Office Review of Starlink (part of RAL review). Reported Jan 94. ||
|| || SP (Starlink Panel) replaced SUC. ||
|| 1994 || Starlink Software Survey (Warren-Smith) (SGP/43). ||
|| || ADAM Review Panel (Hills). Reported Mar 95 (SGP/45). ||
|| || Unsuccessful bid from RGO to relocate Starlink (part of OIM review). Reported Jan 95. ||
|| 1995 || Move to Unix completed (Apr). ||
|| || Starlink Review 1995 (Dufton) - ROSO (Review of Starlink Operations). Reported Oct 95. ||
|| 1996 || Starlink Software Review (Starlink Panel) - efficiency. ||
|| || Starlink Software Survey 2 (Anne Charles). ||
|| || Support for PCs began (Linux). ||
|| || First issue of Starlink software on CD (for PC/Linux). ||
|| || Starlink Software Review (Starlink Panel) - wide ranging. ||
|| 1997 || Starlink Prior Options Review (Robinson). Reported Jan 97. ||
|| || All Starlink software now released bi-annually on CD. ||
|| || ACP (Astronomical Computing Panel) replaced SP. ||
|| || HAC (Head of Astronomical Computing) appointed. ||

The Starlink Chronicle is rather long so it is divided into sections.

  • Site openings lists when each Starlink node was opened.

  • The dates when each node switched from VAX/VMS to Unix.

  • The milestones of the Starlink Project.

The detailed chronicle is broken into groups of a few years.

Year 1977

Jun

Meeting at AAO (Wallace, Bothwell, Peterson, Fosbury, Dickens) to discuss how analysis of astronomical data should be organised. This led to Disney proposing a "Centre for Optical Data Analysis" at RGO. The hardware recommended was based on an Interdata 70 or 7/16 computer with a 64K byte memory and 2x2.5 Mbyte discs. The software would be based on the AAO SDRSYS spectral analysis package. A "low grade programmer" would be required to demonstrate and advise users. Recommended funding was 40K pounds capital and 10K pounds running costs per annum. Ten years later, these initial estimates of the resources required had been exceeded in practice by the following factors: Memory - 2500, Disc - 4000, Capital expediture - 100, Running costs - 200.

Year 1978

Feb 7

Astronomy II Committee considered proposals for image and data processing facilities at ROE, RGO (Hartley), Cambridge (Kibblewhite), and Oxford (A.C. Edwards, Astrophysics). ROE proposed that the Computer Application Services of Heriot-Watt University create a European Image Processing Centre.

Apr 25

Panel on Astronomical Image and Data Processing set up by the Astronomy II Committee of the Science Research Council under the chairmanship of Professor M J Disney.

May 16

Panel endorsed by the Astronomy Space & Radio Board. Observers added from the Astronomy I and Solar System Committees, and the Department of Industry. Terms of reference: "to ascertain the needs of UK astronomers for data reduction and analysis facilities over the next 5-7 years in relation to existing and anticipated facilities, and to report to the Astronomy II Committee".

Year 1979

Feb 27

Meeting at Appleton Lab set up a working group to specify the operational requirements for the Starlink Image Processing System (Lawden, Kibblewhite, Ellis, Shortridge, Davey, Pavelin). Lawden wrote the OR for the computer and Hartley wrote the OR for the image-display system.

Mar

Tenders for computers invited from DEC, GEC, IBM, ICL, Prime.

Apr

Dave Horn (Cambridge) supplied benchmark programs to assess computers. Panel issued its report.

May 15

DEC presentation.

18

IBM presentation.

22

PRIME presentation.

23

ICL presentation.

Tenders received for computers

Jun 18

8th meeting of Panel accepted the DEC tender rather than the Prime (Disney, Aleksander, Alty, Miller, Francis, Terrett, Kibblewhite, Mackay, Davey, Hartley, Lawden, Pavelin, R Martin, Casey, Dean, Ward).

Jul 18

Council approved recommendations for Starlink at a cost not exceeding 1.8M pounds, on condition that the necessary SRC manpower could be made available within the normal complement. Four major centres (ROE, RGO, UCL, Cambridge); two minor centres (RAL, Manchester). All were based on DEC VAX 11/780 with FPA. Major centres had 2 Mbyte memory and 704 Mbyte disk; Minor centres had 1.5 Mbyte memory and 352 Mbyte disk (based on 176 Mbyte RP06). Capital Cost: 1.686M pounds.

19

Abridged Panel report issued by its secretary P J Casey.

Sep 12

SAG/1: 1st Meeting of Starlink Scientific Advisory Group at RL (Disney, Boksenberg, Brand, Burge, Ellis, Kibblewhite, Murdin, Skinner, Hunt, Kemp, Houghton). Project to be run from Computing Division, Rutherford Laboratory. Pavelin to act as Project Manager until a Head of Starlink is chosen.

Oct 8

SAG/2: Delays in SIGMA demonstrating their ARGS caused dismay: "No further time could be given to SIGMA and it was agreed that the order (for image-display hardware) should go to Grinnell."

25

SMM/1: 1st Site Managers Meeting - RAL

31 - Nov 1

Starlink Software Workshop at Appleton Laboratory (Allen, Crane, Fosbury, Golton, Hartley, Hewer, Kibblewhite, Lawden, Murdin, Norris, Pavelin, Skinner, Snijders, Wallace, Warne, Wells). The recommendations of this workshop laid the basis for Starlink's software development plans and the initial design of the Starlink Software Environment.

20

1st Starlink Progress Meeting (Pavelin, Walter, Lawden, Claringbold). Draft outline low-level interface specs produced: timetable - Feb 80: draft software specs, Apr 80: final software specs, Aug 80: installed.

VAX 11/780 delivered to RAL. Grinnell image display configuration agreed but authorisation is required from Dept of Industry. DOI wanted Starlink to buy SIGMA in spite of delays. Two "Image Processing Workstations" were envisaged per site, to consist of command terminal, graphics terminal and image display.

Lorna Claringbold joined Starlink at RL as operator/manager.

Anton Walter joined Starlink at RL as systems programmer.

Grant money approved by Solar System Committee for IPIPS should be added to the Starlink budget and administered within Starlink.

Dec 10

SAG/3: Outline programme for a "software environment" accepted.

RAL VAX working.

Year 1980

Jan

ROE, RGO VAXs delivered.

Many sites expressed dismay at the prospect of Versatecs only, without lineprinters. It was hoped that sufficient funds would be available for a cheap printer at each site.

Order placed for SIGMA ARGS displays.

Bob Dickens joined Starlink as Head of Starlink (half time).

RAL Estimates (80/81) and Forward Look:

DMY

8

8

8

7

6

6

Staff costs/overhead

114

Recurrent

351

Capital

650

Total

1115

461

527

529

502

440

Feb 6

Outline specification of Starlink environment issued (SAG/3/80).

11

SAG/4: Most sites felt a lineprinter was needed. Site managers approved - Dave Terrett (MAN), Sid Wright (UCL), Dave King (RGO), Carol Robinson (CAM).

14

SMM/2 (RAL)

CAM, MAN VAXs delivered.

Mar 4

Dave Pearce joined Starlink at RL as assistant to Lawden.

22

Draft application programmer environmental description issued.

28

1st Starlink Software Release.

Bridget Astridge (Starlink secretary RAL) left Starlink.

31

Bob Fosbury became the first registered Starlink `User' when he signed up at RGO (6 weeks before Barry Kent - claimed as first user in Bul 3).

Apr 1

Starlink started as a standard project within Computing Division, RL.

Starlink officially began offering a service to users.

15

Dave Pearce moved to work on System Software under Walter.

28

SAG/5: All 6 VAXs delivered; not yet running at RGO and MAN.

Starlink Bulletin 1 (ENTERPRISE) issued.

Slavka Cole joined Starlink part time as Starlink Secretary RAL.

May 28

SMM/3 (ROE)

1st SIGMA ARGS image display device installed at RAL.

Jun 27

SAG/6: Bernie McNally now Site Manager at ROE. Local Management Committees set up.

Jul

Starlink Bulletin 2 issued.

VAX installation at MAN completed - last of the original 6 sites.

Sites started to communicate via DECNET at 2.4 Kbaud. Remote file access was the only facility supported.

Initial Site Managers were: CAM - Carol Robinson MAN - Dave Terrett RAL - Lorna Claringbold RGO - Dave King ROE - Bernie McNally UCL - Sid Wright

Aug

Pavelin & Wallace attended `Image Processing in Astronomy' conference at JPL, and visited Kitt Peak, VLA, NRAO (Charlottesville), DEC (Boston).

Sep 1

Patrick Wallace joined Starlink at RL as Head of Applications Software.

4

SMM/4 (CAM)

23

1st release of INTERIM software environment (first Starlink environment).

29

Meeting to discuss End-User Environment (based on draft SGP/6 produced by Lawden). Decided that command language is to be designed by Walter & Wright (monitored by Shortridge). Timescale - Nov 80: Technical proposal produced; Dec 80: Primer presented to community; Feb 81; Ready.

Office moves at RAL caused by opening of R31. People have offices in R27 and R31.

Oct 6

Paper on "The development of Starlink Applications Software" issued.

9

RAL Estimates (81/82) and Forward Look:

DMY

8

8

8

7

6

6

Staff costs/overhead

112

112

112

98

84

84

Recurrent

333

350

355

342

392

378

Total

445

463

467

440

476

462

20

SAG/7: VMS 2.0 installed on RAL VAX. Decided not to buy Eigen video system.

24

Grand Inauguration of Starlink at RAL by Mr D N MacFarlane, Parliamentary Under Secretary at the Dept of Education and Science.

Nov

Starlink Bulletin 3 issued, together with `Glossy Brochure'.

Dec 3

SMM/5 (RAL)

Starlink: Chronicle (last edited 2009-09-15 17:15:45 by MalcolmCurrie)