CERN 29-Apr-99 Computer architectures and operating systems Is there an opportunity for convergence? FOCUS meeting CERN - 29 April.

Slides:



Advertisements
Similar presentations
Technology Analysis LINUX Alper Alansal Brian Blumberg Ramank Bharti Taihoon Lee.
Advertisements

CERN december-00 La Politique du tout PC au CERN Séminaire EDF Clamart – 15 décembre 2000 Les Robertson CERN/IT – Genève.
CERN juin-00 La Politique du tout PC au CERN CUIC – Arcachon – juin 2000 Les Robertson CERN/IT – Genève.
Embedded System Lab. What is an embedded systems? An embedded system is a computer system designed for specific control functions within a larger system,
Amazon. Cloud computing also known as on-demand computing or utility computing. Similar to other utility providers like electric, water, and natural gas,
Industry Classification Computer Hardware and Related Software iMac MacBooks Mac Pro Desktops iPad Mac Displays Time Capsule Xserve/ Mac Pro Servers Mac.
'08 Rabat Why are we using FreeBSD? Scaleable Services Workshop AfNOG 2008 Rabat, Morocco slides by Hervey Allen presented by Joe Abley.
Password?. Project CLASP: Common Login and Access rights across Services Plan
Operating Systems Parallel Systems (Now basic OS knowledge)
Introduction to UNIX Acknowledgement:Thanks to Dr Andrew Horner for the original version of this set of slides. All trademarks are the properties of their.
Web Server Hardware and Software
Server Platforms Week 11- Lecture 1. Server Market $ 46,100,000,000 ($ 46.1 Billion) Gartner.
Complementary Goods Suppose that you need to have both left and right shoes. When you have both left and right shoes, your satisfaction is 10. What is.
Xuan Guo Chapter 1 What is UNIX? Graham Glass and King Ables, UNIX for Programmers and Users, Third Edition, Pearson Prentice Hall, 2003 Original Notes.
Chapter 4 - Software – Part 2 Dr. V.T. Raja Oregon State University.
Introduction to UNIX Acknowledgement:Thanks to Dr Andrew Horner for the original version of this set of slides. All trademarks are the properties of their.
Server Operating Systems Last Update Copyright Kenneth M. Chipps Ph.D. 1.
AN INTRODUCTION TO LINUX OPERATING SYSTEM Zihui Han.
Software Developer Career. ◦ Desktop Program development ◦ Web Program Development ◦ Mobile Program Development.
CredoGov VDI Introduction James Gunn
Abstract Load balancing in the cloud computing environment has an important impact on the performance. Good load balancing makes cloud computing more.
14th April 1999Hepix Oxford Particle Physics Site Report Pete Gronbech Systems Manager.
Chapter 10 – UNIX. History In late 1960s, two employees of Bell Labs (Ken Thompson & Dennis Ritchie) designed a new operating system to overcome the constraints.
CS110/CS119 Introduction to Computing (Java)
IBM Governmental Programs Open Computing, Open Standards and Open Source Recommendation for Governments.
The PC The PC is a standard computing platform, built around a EISA bus (1988) –IBM compatible –“Intel Architecture” from Intel or AMD or other companies.
Open Source Software An Introduction. The Creation of Software l As you know, programmers create the software that we use l What you may not understand.
The evolution of HR David Woodward Chief Information Officer.
CERN 25-Mar-99 Computer architectures and operating systems How many do we have to support in HEP? HEPCCC meeting CERN - 9 April.
Java Beserkers Group 4. Start of Java Development began on June of 1991 by a group of computer scientist at the Sun Mircrosystems Company Development.
Silberschatz, Galvin and Gagne  2002 Modified for CSCI 399, Royden, Operating System Concepts Operating Systems Lecture 1 Introduction Read:
Module 1 Introduction to UNIX/Linux
October, Scientific Linux INFN/Trieste B.Gobbo – Compass R.Gomezel - T.Macorini - L.Strizzolo INFN - Trieste.
1 University of Palestine Faculty of Business Administration and Finance M.I.S Applications Application of Information Systems (Linux) Prepared By: Muhannad.
© 2004 Oracle Corporation Laurent Sandrolini Vice President Systems Platform Division Oracle Corporation.
WINS Data provides customized software solutions, web development, client-server applications using all major up-to-date technologies. We have experience.
June 25, 1998 Is NT needed at CERN? Andreu Pacheco, LHCb / CERN 1 LHCb Computing Architecture Meeting Is NT needed at CERN? Andreu Pacheco LHCb / CERN.
CERN IT Department CH-1211 Genève 23 Switzerland t Experience with Windows Vista at CERN Rafal Otto Internet Services Group IT Department.
Nic Shulver, Using Operating Systems What is a PC?
13 th May 2004LINUX, which LINUX?1 Presentation to the AB/CO Technical Committee – Linux as the Future Console O/S Alastair Bland, 13 th May 2004.
Unix Background. Introducing Unix Brief Unix History u In 1969, Ken Thompson at AT&T Bell Labs began developing Unix. –First done in assembly language.
Open Source & Freeware Products in Maine State Government B. Victor Chakravarty October 6, 2008.
CS 390 Unix Programming Summer Unix Programming - CS 3902 Course Details Online Information Please check.
Mr C Johnston ICT Teacher BTEC IT Unit 06 - Lesson 03 Choosing A Programming Language.
Status of Windows 2000 deployment at CERN Alberto Pace, for the IT/IS group - April 2002
Using Virtual Servers for the CERN Windows infrastructure Emmanuel Ormancey, Alberto Pace CERN, Information Technology Department.
Robert Crawford, MBA West Middle School.  Explain how the binary system is used by computers.  Describe how software is written and translated  Summarize.
Database Architectures Database System Architectures Considerations – Data storage: Where do the data and DBMS reside? – Processing: Where.
22nd March 2000HEPSYSMAN Oxford Particle Physics Site Report Pete Gronbech Systems Manager.
® IBM Software © IBM Corporation IBM Internal Use Only--Not to be shared outside the company until July 25, 2006 Processor Value Unit Licensing for Middleware.
Test Results of the EuroStore Mass Storage System Ingo Augustin CERNIT-PDP/DM Padova.
March 20001© Dennis Adams Associates 10 False Reasons for using LINUX Or “Beware of Jumping on the Bandwagon” Dennis Adams.
IDE disk servers at CERN Helge Meinhard / CERN-IT CERN OpenLab workshop 17 March 2003.
Cluster Configuration Update Including LSF Status Thorsten Kleinwort for CERN IT/PDP-IS HEPiX I/2001 LAL Orsay Tuesday, December 08, 2015.
14 th April 1999CERN Site Report, HEPiX RAL. A.Silverman CERN Site Report HEPiX April 1999 RAL Alan Silverman CERN/IT/DIS.
1 Virtual Appliance Marketplace MGP Competitive Strategy in Network and Information Industries Amit Bhatnagar Dave Lasley David Wong.
Chapter 1 Basic Concepts of Operating Systems Introduction Software A program is a sequence of instructions that enables the computer to carry.
1 july 99 Minimising RISC  General strategy - converge on PCs with Linux & NT to avoid wasting manpower in support teams and.
MSC.Linux Dr. Stefan Mayer European Technical Sales Manager, Linux Division MSC.Software GmbH, Munich.
LHC Computing – the 3 rd Decade Jamie Shiers LHC OPN meeting October 2010.
Virtual Server Server Self Service Center (S3C) JI July.
The Case against Microsoft. © 2004 Pearson Addison-Wesley. All rights reserved12-2.
Peter Idoine Managing Director Oracle New Zealand Limited.
Red Hat, Inc. The Revolution of Choice. Red Hat, Inc. Founded in 1995 –Bob Young, CEO - Co-founder –Marc Ewing, CTO - Co-founder Headquartered in Research.
10-Feb-00 CERN HepCCC Grid Initiative ATLAS meeting – 16 February 2000 Les Robertson CERN/IT.
Operating System & Application Software
Why PC Based Control ?.
CERN openlab for DataGrid applications
O.S. Security.
Presentation transcript:

CERN 29-Apr-99 Computer architectures and operating systems Is there an opportunity for convergence? FOCUS meeting CERN - 29 April 1999 Les Robertson CERN/IT

CERN 25-Mar-99 - #2les robertson - cern/it Summary  A claim that PCs with Linux and Windows are enough for all of HEP’s needs  An assorted collection of caveats  A proposal for a convergence policy

CERN 25-Mar-99 - #3les robertson - cern/it Claim: For the foreseeable future (5 years)  PC hardware  cheapest, best price/performance, as fast as is needed  reliable, real commodity  Windows  a 150M unit market  masses of productivity tools, applications  Linux  by far the biggest ever Unix  the standard for Intel architecture  preferred physics environment  legacy applications & experience  Hard times for the RISC vendors

CERN 25-Mar-99 - #4les robertson - cern/it Conclusion  PCs with Linux and Windows are just fine  price  reliability  usability & functionality  support  There is no need to support anything else

CERN 25-Mar-99 - #5les robertson - cern/it March 97 - first PCs for physics batch April PCs (400 processors)

CERN 25-Mar-99 - #6les robertson - cern/it What would we lose? (the other side of the coin) RISC hardware and software provide -  Solid, reliable hardware, designed to maximise performance  Industrial strength operating systems  clear functionality spec  performance-tuned  multi-processor tested  deterministic support, with clear escalation process  Certified hardware + software combinations  Focus on system (as opposed to component) design  and willingness to discuss product strategy, requirements, joint developments, etc. with HEP labs (or is that just a marketing ploy?)  The possibility to ensure program portability necessity

CERN 25-Mar-99 - #7les robertson - cern/it Are Java stations an option?  SUN, Oracle, etc. propose light-weight Java stations  Really low cost, with zero administration.. but everything has to be written in Java  Good for naturally server-centric applications (e.g. database applications, Web)  Not so obvious for applications that do not have a natural client-server model, and/or use multiple languages (like most HEP applications)  low cost Java stations will keep up the pressure on Wintel  but otherwise, for HEP applications and HEP users …… NO like games machines

CERN 25-Mar-99 - #8les robertson - cern/it Is Linux ready for HEP?  Linux and the rest of Free Software offer:  high quality design  excellent functionality  the products of massive intellectual investment  a sustainable environment  The functionality as a desktop system, or simple server are fine ……… but many proprietary Unix systems have much better server functionality ……… however, the evolving farm models place limited requirements on data servers  Linux tape and disk server evaluations at CERN look promising

CERN 25-Mar-99 - #9les robertson - cern/it Is Linux ready for HEP?  Support is still a problem  especially for load-related, configuration-related issues on servers But there is a rapidly growing number of commercial offerings (including IBM+Red Hat, HP+Red Hat) and HEP could probably justify a kernel programmer or two (since there is no licence fee for the OS)  Lacking third party products  or well-supported ports of third party products (and in particular Objectivity)  changing fast - but this is clearly still a concern

CERN 25-Mar-99 - #10les robertson - cern/it Is Linux ready for HEP? Not quite for all applications, but the situation is changing very fast

CERN 25-Mar-99 - #11les robertson - cern/it Are we sure that Linux is the right choice for Unix on Intel?  No - we cannot be sure there will be competition from the suppliers of proprietary Unix (Compaq, IBM, HP) Linux is the clear and growing leader today - it will be hard to beat

CERN 25-Mar-99 - #12les robertson - cern/it Do we also need Windows?  The majority might say No  Windows is hard to use for physics batch, especially when we try to make it look like Unix  It probably is not much better (maybe even worse) as a server than Linux  Support for complex environments is no clearer than with Linux  But there are so many excellent “productivity” tools and applications sustained by the 150 million unit Windows market  How can people get by without Developer’s Studio?  How can Star Office keep up with MS Office?  ……….  How can we ignore the biggest ever operating system?

CERN 25-Mar-99 - #13les robertson - cern/it An opportunity for standardisation in HEP  At CERN (& in general HEP) we are running far too many operating systems, in many cases with only slightly different functionality or hardware cost benefits - - and at a high cost for users and support teams  PCs + { Linux ¦ Windows } offer an historic opportunity to reduce the solution set  The requirements of LHC computing -  massive numbers of processors/boxes  integration of regional computing centres and CERN demand a common agreement on what will be supported requirement

CERN 25-Mar-99 - #14les robertson - cern/it Proposal for a policy  Restrict ourselves to PC hardware (IA32 -> ? IA64)  with Linux or Windows 2000  Beef-up investment in Linux and Windows  bring support up to the standards of proprietary Unixes  tackle the problems of scaling the management and performance of desktops and physics farms  Develop a migration plan -  progressively freeze support for other Unixes, announcing end-dates which are reasonable for old experiments,  but discourage strongly further investments in RISC systems by current and future experiments  Try to have the same policy adopted in the HEP community

CERN 25-Mar-99 - #15les robertson - cern/it Just to be clear ----  This is a proposal for a convergence policy  which looks realistic now  and will provide a single starting point for LHC computing  but we can be sure that the business will not stand still, and we shall sooner or later have to expand the systems and architectures supported AIX WNT Irix Solaris Digital Unix HP-UX MAC-OS Linux Windows 95 SPARC MIPS Intel IA-32 PA-RISC Power PC Alpha Linux Windows 2000 Intel IA-32/ ?

CERN 25-Mar-99 - #16les robertson - cern/it Conclusion  PCs, Linux and Windows 2000 enable a major simplification of the environments supported at CERN, where our needs mostly coincide with those of many other users  We can and should take the decision now to standardise on these  following and stimulating the mass market in most cases  restricting HEP specials to really, really special issues  building up very good support & expertise  progressively freezing support for other systems, architectures