- Physics-Departmental-Computing

Computing in teaching support meeting 15:00 today.
by David Colling 10 Jan '23

10 Jan '23

Dear All, Sorry that there is no ideal time and thanks to those of you who have moved other meetings etc. First the coordinates. I am only expecting a handful of people so anybody in college can come to my office (Blackett 505) and I will be using the VC set up there. Otherwise the teams meeting will be at: https://teams.microsoft.com/l/meetup-join/19%3aa4aacd8a744e4aeead6f43e3da2c… Background =========== We use computers for teaching, both in dedicated computing suites and in experimental labs. Computing Teaching =================== - We have 120 machines in a dedicated suite on level 3 and these are used to teach the students how to program (mainly in python). These are quite often in a mess - machines on the floor (about 25% of them this morning), keyboards in a pile at he side of the room and sometimes/often in a poor software state such that (for example) anaconda cannot be installed for use by the students. The default solution when this happens is to tell them to "try another machine". We need to decide who is responsible for the support of these machines and how we do it. Leigh previously promised ICT support, but it doesn't appeared to be working (or happened at all). - Need plagiarism suite for code. Have been using MOSS but that is now slow and overworked. Were promised Gradescope (which has a commercialised version of MOSS inbuilt) at the beginning of the year, but this has failed to happen so far. This is needed. - Looking to next year, we would like to automate so much of at least the second year computing, by getting the students to use git and to use code checking tools directly both in their IDE and in git. This needs to be built now for next year. - Review of software installed. Computers in teaching labs =========================== - We have computers in several teaching labs with different configurations. These are mostly windows but there is some linux. The problems that with these set ups are mainly the lack of timely support and this is compounded by the lab technicians not having admin rights (would not be needed if support was fast). Last year, we were 12 hours away from cancelling an UG course because the correct software (LabView) was not installed on the machines that it was going to use despite a ticket being raised to install the software a month in advance. We have other machines were a needed package (Geant4) runs most of the time for most people but not all of the time for all of the people. Again the solution here seems to be to "try another machine". - Looking to the future, we would like to move a to a world where there are series of standard build for the different labs that work for all students. - Continued support for Linux in labs. Anything else? ================ -????? - Project support? Best, david

1 0

Meeting regarding support for teaching computers in Physics.
by David Colling 09 Jan '23

09 Jan '23

Hi All, On my list of actions from opur discussion on Friday was to set up a meeting to try to sort out the support for computers used in teaching in Physics. These include those in the computing suite and those in the different labs. I was going to try to set up a doodle poll for this, but my next 2 weeks are extra ordinarily busy (or this become ordinary these days?). So I am going to suggest a day and time and if the key people cannot make it then I will set up a doodle. So I am going to suggest 15:30->16:30 on Wednesday (11/1/2023). This will hopefully be a mixed meeting. While everybody is welcome, I think that the key people I think include Okan, Graham from the UG labs, at least one person from either first or second year computing (ideally both but ...). It would also be good if anybody else teaching a course that is heavily computer based can make it and perhaps Luke if he is available. I have a list topic to discuss but please let me know if you have suggestions. Best, david

2 4

Fwd: Possible Collaboration with Intel and Lenovo
by David Colling 06 Jan '23

06 Jan '23

This is the mail that I sent to the staff list. As promised I am sending to this list as well as not everybody is on the staff list. -------- Forwarded Message -------- Subject: Possible Collaboration with Intel and Lenovo Date: Fri, 6 Jan 2023 09:39:54 +0000 From: David Colling <d.colling(a)imperial.ac.uk> To: ph-staff-dl <ph-staff-dl(a)imperial.ac.uk> Dear All, Imperial has entered into a partnership with Intel and Lenovo. While this is largely over the provisioning of HPC systems at Imperial it is also meant to encourage joint research ventures. This has been given the somewhat cheesy name of ICICLE (Imperial College Intel Corporation LEnovo). Intel are making an upfront investment of $250,000 in joint projects, however in our initial meeting with them, they pointed out that when they have done this with similar organisations in the past, this figure has quickly grown to a substantial number of $M. So I think that there is potential for significant collaboration here if your research fits with Intel's interests. I have attached some slides from the initial meeting and slide 3 outlines the 6 areas in which Intel are interested in collaborating. Please read and contact me if you think that your research interests fit into any of these areas. The next meeting with Intel and Lenovo is next Wednesday (11/1/2023) morning so it would be helpful if you have ideas that you send them to me by (say) midday on Tuesday. Best, david

1 0

Informal Computing chat tomorrow at midday
by David Colling 05 Jan '23

05 Jan '23

Hi All, As I said in my previous email I suspect that there won't be many people around tomorrow, but do want us to meet if only informally before the FRCC next Wednesday. For those around in college tomorrow I have booked Blackett 532 and for those who will be remote there is a teams meeting booked in the appropriate team and the link for this is: https://teams.microsoft.com/l/meetup-join/19%3aa4aacd8a744e4aeead6f43e3da2c… The items that I would like us to discuss are: - Update from ICT - including HPC refresh status and other items (Led by Ellen and Andy). Not to include managed devices (next item). - Managed devices - we all know about this but there was some progress before Christmas which I think will be useful (led by Ingrid). Ingrid, if there is a current version of the document tht we discussed before Christmas then if you send it to me I can upload it to the teams site. - The management of teaching computers in UG labs and computing suites. This is something that we have never really satisfactorily resolved and with Leigh having left we might be able to make some progress on this. I hope that tomorrow we can start to set out a road map as to how we build a solution in this area. - RSEs - discussed several times before, but with a planning round approaching I want us to make progress on this and get something in this year. - Possible physics involvement in collaborative projects with Intel and Lenovo. - (only if we have time) FAIR data - continuation from previous discussions. I will upload the document sent to Mary Ryan to the teams area. This is to be discussed at RC on 14th of Feb. - (probably not enough people but) how do we organise the timings of these meetings in future. - AoB Best, david

1 0

"Blwyddyn Newydd dda a" date of next meeting
by David Colling 04 Jan '23

04 Jan '23

Dear All, I hope that you are all having a wonderful break - I am - and I apologise for sending a work email in the middle of it. As you know we tend to have a Physics Computing Meeting in the week leading up to the the FoNS Researsh Computing Meeting (FRCC). The next FRCC is on 11th Jan 2023 so I have set up a doodle poll for the week leading up to this. As you know we tend to have the Physics Computing meetings at lunch time. The doodle poll is at: https://doodle.com/meeting/participate/id/dJy4zP9b I have invited Ingrid because one of the items I want on the agenda is managed devices. Nothing has caused so much consternation amongst the user community for many years but I think that we are reaching a good conclusion. Please send me suggestions for other agenda items - I have some in mind but I am sure that there are others. One last thing, something that Andrew brought up at the end of the last meeting. We know the dates of the FRCCs for the rest of the academic year. In the past I have have put up doodle polls for the dates of our meetings only a week or two in advance of the FRCC. My thinking was that people would best know their availability at that stage. I could easily either put up doodle polls now for future meetings or even simply set the dates of the Physics meetings (say) 1 week before the FRCCs. The reason that I haven't done so in the past is that those of us in large collaborations often have meetings that cause clashes introduced at a few weeks notice, however these appear to still cause problems even with the approach that I current approach. Anyway, I would welcome people's opinions as how best to organise things in the future. Happy New Year to all! Best, david

1 2

Re: [Physics-Departmental-Computing] Services needed on a research computing desktop and laptop
by Jaffe, Andrew H 09 Dec '22

09 Dec '22

Dear all, Apologies for this slightly spammy follow-up, but this seemed like a good group to ask about this: Until what I think was a few weeks ago, I was able to use ssh and screen-sharing (vnc) to access my machine at Imperial, on the wired network with a fixed host name and IP address. From on-campus I could do this directly, and from off campus it required the vpn. It was also possible to use ssh via the sshgw machine from off campus even without vpn. I had successfully done this from other macOS machines, as well as iPads and iPhone with appropriate ssh and screen-sharing software. Recently, all of these modes of access have become impossible. I do not think this is due to any changes on my machine itself. I have successfully been able to ssh from the machine to itself, but actually using its IP address, which I think at least means that it is accepting incoming ssh. But user error on my part certainly remains a possibility… Is anyone else experiencing this? Is it a known change? Is it related to Unified Access (Zscaler), which as far as I know has not yet been deployed? Did I miss a communication about this? (This page < https://www.imperial.ac.uk/admin-services/ict/self-service/connect-communic… <https://www.imperial.ac.uk/admin-services/ict/self-service/connect-communic…> > is unclear about whether any of this is currently possible.) Many thanks, Andrew ______________________________________________________________________ Professor Andrew Jaffe a.jaffe(a)imperial.ac.uk<mailto:a.jaffe@imperial.ac.uk> Astrophysics Group +44 207 594-7526 Blackett Laboratory, Room 1018B Imperial College, Prince Consort Road London SW7 2AZ UK http://imperial.ac.uk/people/a.jaffe On 30 Nov 2022, at 10:39, Bantges, Richard J <r.bantges(a)imperial.ac.uk> wrote: Hi David (and everyone else), Having consulted with members in SPAT (those that have responded in time), here is our overview of services required on a research class machine: (Below are my definitions of service classes, others may have different ones of course) Admin Services: ICIS Authentication Services: LDAP, Licence Servers (e.g. Matlab, IDL, etc.) Communication Services: SSH (typically Port 22), FTP (?), Zoom, Access to Remote Data centres, Remote Desktop Gateway, Unified Access, VPN Data Sharing Services: SMB Servers, HTTP / HTTPS servers, NFS Servers, File Exchange (retired in 2 days?), Sharepoint, Onedrive, etc. Information Services: DNS, Library Journals Productivity Services: Office365, HPC Interactive Services (e.g. Jupyter Notebook, etc.), Software Centre Teaching Services: Blackboard, Starfish, Panopto, Secure College Webpages The above is probably a non-exhaustive list, and I've no doubt once a clearer picture emerges of what a Research Class machine might look like, then there will be the hope that this can be iterated / fine-tuned to accommodate any niche requirements inherent with the breadth of research activities and their associated diverse requirements. Thanks for your efforts towards defining this David. Best wishes, Rich -----Original Message----- From: David Colling <d.colling(a)imperial.ac.uk> Sent: 22 November 2022 20:17 To: Bresme, Fernando <f.bresme(a)imperial.ac.uk>; French, Paul (PHOT) M W <paul.french(a)imperial.ac.uk>; Keaveny, Eric E <e.keaveny(a)imperial.ac.uk>; Sternberg, Michael J E <m.sternberg(a)imperial.ac.uk>; Staffell, Iain L <i.staffell(a)imperial.ac.uk>; Pengelly, Ellen <e.pengelly(a)imperial.ac.uk>; Buchaca-Domingo, Ester <e.buchaca-domingo(a)imperial.ac.uk>; Bantges, Richard J <r.bantges(a)imperial.ac.uk>; Michalickova, Katerina <k.michalickova(a)imperial.ac.uk>; physics-departmental-computing <physics-departmental-computing(a)imperial.ac.uk>; Bryce, Craig T <c.bryce(a)imperial.ac.uk>; Bearpark, Michael J <m.bearpark(a)imperial.ac.uk> Cc: David Colling <david.colling(a)gmail.com>; Pearse, Will <will.pearse(a)imperial.ac.uk>; Cucinotta, Clotilde <c.cucinotta(a)imperial.ac.uk> Subject: Services needed on a research computing desktop and laptop Hi All, I am sending this to the Physics Departmental Computing Committee and to the departmental members of the FRCC so that they can gather information from their departments. As some of you know ICT are increasingly confining what people can do on college machines, even those bought on research grants and used by individual researchers. This has been most noticed by the change in the management of Macs. In my years involved in departmental computing, no issue has annoyed more people. Behind this is the increased number of attacks on university computing system which is visible both at Imperial and elsewhere. Some universities have been badly hit and have ended up paying £Ms to ransomware attackers. Apparently this is one of the things that keeps our President awake at night. This is clearly a threat that we have to take seriously, but it is also not clear how much damage could be done to college systems by a laptop or desktop used by a single (or team of) researcher(s). In discussions with ICT the most sensible approach seems to be that we define a class of machine that is a research desktop or laptop that ICT don't manage but which also has limited access to college central systems. Most of us have no reason to access payroll (say) and in fact would view it as a security breach if we could. We have a meeting on the 30th November where we will discuss this proposed set up. What I need going into is the list of services that researchers would need access to from these research machines, how that access would them + any other thoughts/comments. For example the sort of thing that occurred to me are: service: Office365 (including sharepoint, email, OneDrive teams etc) Access: Is access through the secure web portal enough for most of these plus a mail client providing secure access the the email. [I use Office365 much less than almost anybody to whom this email is going so am the least qualified to answer this one] Service: ICIS (Payslips, expenses claims etc) Access: Secure web access should be enough. Service: Starfish Access: Secure web access is sufficient. What other services are needed and how? Other comments: - I don't think that it is unreasonable to have a requirement that the disks of all research laptops are encrypted in case they are lost when travelling. The performance hit is minimal and if that is important then running on a laptop might not be ideal. So please do send me your thoughts (on services mainly) and comments. For once I would not be against you sending to everybody as I would welcome debate on this. Best, david _______________________________________________ Physics-Departmental-Computing mailing list Physics-Departmental-Computing(a)imperial.ac.uk https://mailman.ic.ac.uk/mailman/listinfo/physics-departmental-computing

2 1

Feedback needed at short notice
by David Colling 07 Dec '22

07 Dec '22

Sorry for the short notice - completely my fault. Tomorrow afternoon there is a workshop on data storage in the future. The agenda is below and I would welcome any feedback from people. Basic questions/criteria for the workshop (not limited to): Active Data Storage · What volume of data do you need to store during the lifetime of an active research project? · Are you using or planning to use RDS active data storage in conjunction with the RCS HPC resources or for other data storage requirements? · Do you need to share data externally? · Do you need to transfer data in/out of the next generation RDS from external resources? Disaster Recovery · What volume of data do you need to protect during the lifetime of an active research project? · If the data storage platform was significantly affected what ‘time to restore’ could you accommodate? § I need instant access to data § OR Project would require data to be restorable within hours/days/weeks/months? Archive · What volume of data would you need to archive at the end of a research project? · For what time period would you need to retain data in an Archive? · Do you have sensitive data (e.g. personal/commercial/other) that would need to be archived and an archive meet certain information governance requirements?

1 2

Fwd: Fact Finding with ICT
by David Colling 01 Dec '22

01 Dec '22

Please forward to anybody who might be interested... -------- Forwarded Message -------- Subject: Fact Finding with ICT Date: Thu, 1 Dec 2022 09:57:52 +0000 From: Villamil, Juan <juan.villamil(a)imperial.ac.uk> To: Bryce, Craig T <c.bryce(a)imperial.ac.uk>, French, Paul (PHOT) M W <paul.french(a)imperial.ac.uk>, Colling, David J <d.colling(a)imperial.ac.uk>, Bresme, Fernando <f.bresme(a)imperial.ac.uk>, Keaveny, Eric E <e.keaveny(a)imperial.ac.uk>, Sternberg, Michael J E <m.sternberg(a)imperial.ac.uk>, Staffell, Iain L <i.staffell(a)imperial.ac.uk>, Buchaca-Domingo, Ester <e.buchaca-domingo(a)imperial.ac.uk>, Armstrong-Brown, Sophie <s.armstrong-brown(a)imperial.ac.uk>, Bantges, Richard J <r.bantges(a)imperial.ac.uk>, Michalickova, Katerina <k.michalickova(a)imperial.ac.uk>, Li, Yangna <yangna.li(a)imperial.ac.uk>, Pietzuch, Peter R <prp(a)imperial.ac.uk>, Bearpark, Michael J <m.bearpark(a)imperial.ac.uk>, McCann, Julie A <j.mccann(a)imperial.ac.uk>, Cantwell, Chris <c.cantwell(a)imperial.ac.uk>, Bangerter, Neal K <n.bangerter(a)imperial.ac.uk>, Del Rio Chanona, Antonio <a.del-rio-chanona(a)imperial.ac.uk>, Taborda, David M G <d.taborda(a)imperial.ac.uk>, Boyle, David <david.boyle(a)imperial.ac.uk>, Wickerson, John P <j.wickerson(a)imperial.ac.uk>, Gorman, Gerard J <g.gorman(a)imperial.ac.uk>, Angioletti-Uberti, Stefano <s.angioletti-uberti(a)imperial.ac.uk>, Galvan, Stefano <s.galvan(a)imperial.ac.uk>, Wright, Andrew G <andrew.wright1(a)imperial.ac.uk>, Cohen, Jeremy <jeremy.cohen(a)imperial.ac.uk>, Sandhu, Manjinder S <m.sandhu(a)imperial.ac.uk>, Richards, Andrew <a.j.richards(a)imperial.ac.uk>, Prescott-Jones, Steven <s.prescott-jones(a)imperial.ac.uk>, Pengelly, Ellen <e.pengelly(a)imperial.ac.uk> CC: Scholfield, Liz <e.scholfield(a)imperial.ac.uk>

1 0

Fwd: FoNS Device Management - Defining a research devices and user group
by David Colling 01 Dec '22

01 Dec '22

FYI ... Please feel free to send comments ... I have a very busy up to the meeting this afternoon and will not have really enough time time to scrutinise the attached (so far I have not even had a chance to open it) so your comments will be very useful. Best, david -------- Forwarded Message -------- Subject: FoNS Device Management - Defining a research devices and user group Date: Tue, 29 Nov 2022 18:22:48 +0000 From: Joannou, Ingrid <i.joannou(a)imperial.ac.uk> To: Hanham, Neil A <n.hanham(a)imperial.ac.uk>, Prescott-Jones, Steven <s.prescott-jones(a)imperial.ac.uk>, Taylor, James A <james.a.taylor(a)imperial.ac.uk>, Willson, Thomas H <t.willson(a)imperial.ac.uk>, Colling, David J <d.colling(a)imperial.ac.uk>, Bresme, Fernando <f.bresme(a)imperial.ac.uk>, Jaffe, Andrew H <a.jaffe(a)imperial.ac.uk>, Harrison, Nicholas M <nicholas.harrison(a)imperial.ac.uk>, Bearpark, Michael J <m.bearpark(a)imperial.ac.uk> CC: David Colling <david.colling(a)gmail.com>, Pengelly, Ellen <e.pengelly(a)imperial.ac.uk>, Richards, Andrew <a.j.richards(a)imperial.ac.uk>, Villamil, Juan <juan.villamil(a)imperial.ac.uk> Dear All, Apologies for the late sending of this. Attached is a draft proposal based on some of the discussion so far. We are hoping this can be the basis for our conversation tomorrow with the aim to define further that middle layer of research machine. David, please feel free to forward to anybody else that you think might need to be included in this discussion. Kind Regards, Ingrid Dear All, I hope you are well. As per our previous meeting, I am setting up this meeting so we can follow up on defining the research devices with services used and user group for FoNS. A few of us will be on site for this meeting so I have added the room number in ICT but also happy to go to you if you can book a room for us? Kind Regards, Ingrid ________________________________________________________________________________ Microsoft Teams meeting *Join on your computer or mobile app*** Click here to join the meeting <https://teams.microsoft.com/l/meetup-join/19%3ameeting_MmFlNzM4YWEtMmRhNy00…> Meeting ID: 328 697 732 543 Passcode: CgLPAv Download Teams <https://www.microsoft.com/en-us/microsoft-teams/download-app> | Join on the web <https://www.microsoft.com/microsoft-teams/join-a-meeting> Learn more <https://aka.ms/JoinTeamsMeeting> | Meeting options <https://teams.microsoft.com/meetingOptions/?organizerId=add04a20-8545-483d-…> | Legal <https://www.imperial.ac.uk/media/imperial-college/administration-and-suppor…> ________________________________________________________________________________

3 3

Services needed on a research computing desktop and laptop
by David Colling 30 Nov '22

30 Nov '22

Hi All, I am sending this to the Physics Departmental Computing Committee and to the departmental members of the FRCC so that they can gather information from their departments. As some of you know ICT are increasingly confining what people can do on college machines, even those bought on research grants and used by individual researchers. This has been most noticed by the change in the management of Macs. In my years involved in departmental computing, no issue has annoyed more people. Behind this is the increased number of attacks on university computing system which is visible both at Imperial and elsewhere. Some universities have been badly hit and have ended up paying £Ms to ransomware attackers. Apparently this is one of the things that keeps our President awake at night. This is clearly a threat that we have to take seriously, but it is also not clear how much damage could be done to college systems by a laptop or desktop used by a single (or team of) researcher(s). In discussions with ICT the most sensible approach seems to be that we define a class of machine that is a research desktop or laptop that ICT don't manage but which also has limited access to college central systems. Most of us have no reason to access payroll (say) and in fact would view it as a security breach if we could. We have a meeting on the 30th November where we will discuss this proposed set up. What I need going into is the list of services that researchers would need access to from these research machines, how that access would them + any other thoughts/comments. For example the sort of thing that occurred to me are: service: Office365 (including sharepoint, email, OneDrive teams etc) Access: Is access through the secure web portal enough for most of these plus a mail client providing secure access the the email. [I use Office365 much less than almost anybody to whom this email is going so am the least qualified to answer this one] Service: ICIS (Payslips, expenses claims etc) Access: Secure web access should be enough. Service: Starfish Access: Secure web access is sufficient. What other services are needed and how? Other comments: - I don't think that it is unreasonable to have a requirement that the disks of all research laptops are encrypted in case they are lost when travelling. The performance hit is minimal and if that is important then running on a laptop might not be ideal. So please do send me your thoughts (on services mainly) and comments. For once I would not be against you sending to everybody as I would welcome debate on this. Best, david

10 28