Here's yet another non-linux post. I'm currently getting ready for the start of the new semester and teaching, and so haven't had much time to work on improving my computer skills.
Anyway.
I've been advertising for an international PhD student for the past 9 months and have so far only had one great applicant and three acceptable applicants. That's out of ca 200 applicants in total.
So what does 'acceptable' mean? In this case my use actually agrees with the literal meaning -- students which will stand a chance of being accepted to the PhD program. It also means students which I could imagine working with.
The formal requirements will likely differ between different institutions, and between supervisors. In addition, some supervisors may be looking for different personalities in their prospective hires, than others.
I don't think that I'm being unnecessarily harsh in evaluating applicants, as I've had colleagues review my shortlists and who have thought I've even been a bit too optimistic in my evaluations.
At any rate, if you are looking for a PhD, be aware that there are a lot of applicants out there, and only a limited amount of money and places, so you will want to spend some time on your application.
So here are a few of my thoughts:
Before reading, keep in mind that I understand that applying for a PhD, especially if you are from the developing world and applying for a PhD position in the industrialised world, can be very tough, and sometimes depressing. You don't receive a reply to most of your applications, and when you do, they responses are normally negative.
* Try to familiarise yourself with the formal requirements, and address them in the first paragraph in your email to a prospective supervisor. In the case of my uni, there are two main requirements:
-- an undergraduate degree equivalent to a first class honour degree in Australia
-- a sufficiently good score at the IELTS
That's it. However, the hubris of many universities in Australia mean that the first requirement is a significant hurdle. Typically, good grades are just the beginning. In addition to that, the applicant needs to hold a masters degree (by research) and have a couple of papers in ISI rated journals. Obviously almost none of our own undergraduate students would meet that, but there you go.
So in your first paragraph, state what unis you did your degrees at, what your cumulative GPAs (or equivalent) were, how many papers you have published and what you overall band score AND section scores on the IELTS (or TOELF) are.
At this stage, that's much more important than your background, your hobbies, or anything else. If you can't meet the minimum requirements for entry to the PhD program, everything else doesn't matter.
* Read the advertisement, and follow any instructions
I ask applicants to submit all their documents as PDFs. Yet, I get plenty of applications with .doc, .docx, jpeg etc attached. You didn't read the instructions -- will you be more careful as a PhD student? Remember that you competing against plenty of applicants that did read the instructions.
Did I ask for your IELTS results? Didn't attach or mention them in your email/CV? Not a good sign. Also, it means that you're probably not a candidate.
* Address the supervisor and the supervisor's research
I get way, way too many emails that start with 'Dear Sir', or 'Dear Professor' or even worse: 'Dear Sir/Madam'. Put my name in there. It'll show me that you spent at least a few minutes on personalising your email. If you don't make that effort, why should I make the effort of reading your email and looking at your documents?
Also, please do mention the research of the supervisor you are applying to. It doesn't need to be anything insightful or special, but just write something like: 'I find your research into catalytic activation of molecules in ionic liquids very interesting.' or 'I read your article in Green Chemistry, 2013, 10, 2345 and found it very interesting. In particular, I liked how it showed how the selectivity of blah blah blah'.
The reason is not that you are showing off your great scientific skills (you've got an undergraduate degree -- we don't expect much), but that it shows you spent a bit of effort writing your email and personalising it. Also, flatter -- in moderation -- can occasionally help (don't go overboard, so be careful -- too much makes you seem insincere).
* Don't cold-call
This should go without saying. I've had one student email me in the morning, then call me in the afternoon. That kind of behaviour is probably correct if you are applying for certain jobs in the Real World (marketing?), but not for a PhD in chemistry. It's a sure-fire way of annoying people.
* Don't send a linked-in invite
I don't have time to scroll through your profile and try to compile a CV for you. Send me your CV in pdf format instead. Also, I don't know you, and have no incentive to add you to my 'network'.
* Be careful about 'hobbies' and 'interests'.
To me as a potential supervisor they really don't matter (again, this is my personal opinion). I know that the idea is to show that you are a well-rounded individual, but knowing that you like 'travel' or that you consider 'internet browsing' a skill will not be the edge that gets you into a PhD programme.
* 'It can't help you, only harm you'.
Keep this in mind. Unless it's a piece of information required in the advertisement, or that you are absolutely certain will help your application, consider leaving it out. You may include it to highlight a particular skill or trait, but remember that a CV can be interpreted ambiguously, and your intent may not be obvious. Instead, what you feel shows how independent and committed you are, can be seen as being unfocussed, a difficult person to work with, or simply attract attention away from more important aspects of your CV.
* Attending lectures, conferences
In their CVs, some applicants include lectures by famous people that they've attended, or conferences that they've gone to.
Here's the problem for me: most first year PhD students struggle with the notion that doing the work is no longer enough. Doing the experiments, or following your supervisor's instructions, is not enough. To get a PhD you need to make that extra effort and making things work. And if it doesn't work, you put in 150% effort -- the extra 50% being extra-curricular work on finding a related project that will work. Life as a PhD student can be easy if you are lucky, but most often is not -- life is incredibly good when you project is working, but on the flip-side it can be hard, depressing and demoralising when it isn't. You supervisor can alleviate some of that, but remember that your supervisor is only there to point you in a general direction -- the PhD is all about making the transition to becoming an INDEPENDENT research.
So be careful -- if you've presented posters or given talks at conferences or at other universities, you should definitely list them, but under a suitable heading -- NOT publications. They'll detract attention from the publications, and the publications is what will get you an offer of acceptance.
* Do not make things conditional
I had an applicant who was borderline (in terms of meeting the requirements), and in those cases occasionally the supervisor putting in extra effort into cajoling the university administration MAY be enough to get a student accepted (don't count on it). If your prospective supervisor asks you to re-take IELTS, don't write something along the lines of 'I will, but only if this is the last hurdle'.
I understand it's expensive, but remember: even if you meet all the requirements I cannot guarantee that you get accepted. And I can' wait months for each student to pass through the application system -- I need to hire someone now. So be proactive.
* Face-to-face (or skype/video) interview is a good sign
If your supervisor asks for a skype interview, this is a great sign. And likely this isn't really done in order to gauge your scientific skills, but just to get a feel for your personality. Also, it's a way of making sure that your English levels are good enough that you can communicate with your supervisor. Finally, if you are borderline in terms of IELTS/TOELF, your supervisor may be able to argue that you English is good enough based on that interview. So take the opportunity.
And send an email a few hours after the interview thanking for the opportunity. 1-2 lines is enough. It will show that you're a decent human being.
* Be prompt in replying to emails
It doesn't matter what stage of the application you are at -- until the paperwork has been signed you are still on probation. If you take several days to reply to any of my emails, then you are likely to be dropped. The reason is simple: if you take a week to get things done when you are a PhD student, then you will be a disaster for me. A disaster that I'll have to live with for the next 3-4 years, and whom will be using up my research grant, and potentially ruining my career.
I understand that the reason for you being slow may be different -- maybe you are just nervous, maybe you have nothing to say, maybe you feel you are intruding. Still, be prompt.
So:
if you can show that you can read and follow instructions, and if you can make my life easy by addressing the selection criteria in a clear way, and if you seem like a person I might enjoy working with for the next 3-4 years, then you stand a fair chance of getting an offer.
If I think you'll need constant supervision, is sloppy and won't follow instructions, or that our personalities will clash, I'll probably avoid you no matter how good your grades are.
Lindqvist - a blog about Linux and Science. Mostly.
Showing posts with label australia. Show all posts
Showing posts with label australia. Show all posts
15 July 2013
15 August 2012
220. My first few lectures in Australia...and the importance of thermodynamics
I'm not an expert on anything. It's not a matter of false modesty, but an observation based on the fact that I come across new things on a regular basis, even within my field of research. When I do, I try to learn.
But to learn new things I need a toolbox - a set of skills that I rely on to put anything new into context. The toolbox for a chemist is basically thermodynamics. Simple, applied thermodynamics. As a chemist you can get away with memorising the expression for Gibbs free energy and the general expression for rate laws, and you can do a whole lot of fun/damage.
I'm only at the beginning of my current class, and I'm an inexperienced lecturer, but I had a bit of a shock today discovering that my class -- second semester of the second year at a 'good' university -- don't feel comfortable dealing with free energies and standard potentials.
The origin lies in the idea of students as clients here -- universities want pass rates in excess of 70-80%, while most of the faculty likely experienced a first year as undergraduates where the Gen Chem class (which did thermodynamics until you eyes were bleeding) had a failure rate of 60% or above.
We might have thought it was harsh to fail that many students at the very beginning, but the result was that the more inspired/motivated students made it through, and the ones who weren't willing to dedicate the effort necessary to become professionals got a kick in the pants to look for majors that actually interested and inspired them.
If something interests you it becomes 'easy' -- either because you instinctually understand it, or more likely, because you simply put in that extra effort to teach yourself.
Instead, the impetus to pass as many students as possible -- and to get 'good' student feedback which will help your promotion -- means that the students are never challenged. 'Difficult' topics are avoided and taught late in the course or ,increasingly, not at all.
Rumour has it that one of the reasons why some Australian universities are adopting a Bologna-inspired model is because they can use the masters section of the education to cover the things the students should have learned as undergraduates --- and thus still produce graduates with the skills that their chosen major indicates that they should have.
It's pretty damning.
The consequence is dire -- some scary example of PROFESSORS -- that is: professionals entrusted with teaching the next generation of scientists and engineers -- in the STEM fields who don't appear to understand basic thermodynamics or more specifically: entropy and the distinction between open and closed systems based on their use of thermodynamics to 'disprove' Evolution. They may be appearing to be capable professionals in every other sense and may well do 'good' research. The individuals may be appearing to do capable research in every other aspect and may be wonderful people, but their use of the 2nd law of thermodynamics as an argument against evolution is just misguided.
Andrew McIntosh -- Professor of Thermodynamics(!) and Combustion Theory at Leeds. Website.
Stuart Burgess -- Professor, Department of Engineering at Bristol. Website
In fact, the list here would presumably include mostly people of a similar persuasion. While I've seen Andrew's and Stuart's writings, I feel comfortable commenting on their opinions, but since I am not as confident about the rest of the people on the list, let's just highlight the fact that it include people from (the universities of) Sheffield, Cambridge, Liverpool and Cardiff.
Or what about this letter: http://www.bcseweb.org.uk/index.php/Main/EstelleMorris
The fact that the signatories mention their affiliations is an obvious way of trying ot use those affiliations to attach significance to their views.
Anyway, here's MC Hawking's take on it: http://www.mchawking.com/includes/lyrics/entropy_lyrics.php
But to learn new things I need a toolbox - a set of skills that I rely on to put anything new into context. The toolbox for a chemist is basically thermodynamics. Simple, applied thermodynamics. As a chemist you can get away with memorising the expression for Gibbs free energy and the general expression for rate laws, and you can do a whole lot of fun/damage.
I'm only at the beginning of my current class, and I'm an inexperienced lecturer, but I had a bit of a shock today discovering that my class -- second semester of the second year at a 'good' university -- don't feel comfortable dealing with free energies and standard potentials.
The origin lies in the idea of students as clients here -- universities want pass rates in excess of 70-80%, while most of the faculty likely experienced a first year as undergraduates where the Gen Chem class (which did thermodynamics until you eyes were bleeding) had a failure rate of 60% or above.
We might have thought it was harsh to fail that many students at the very beginning, but the result was that the more inspired/motivated students made it through, and the ones who weren't willing to dedicate the effort necessary to become professionals got a kick in the pants to look for majors that actually interested and inspired them.
If something interests you it becomes 'easy' -- either because you instinctually understand it, or more likely, because you simply put in that extra effort to teach yourself.
Instead, the impetus to pass as many students as possible -- and to get 'good' student feedback which will help your promotion -- means that the students are never challenged. 'Difficult' topics are avoided and taught late in the course or ,increasingly, not at all.
Rumour has it that one of the reasons why some Australian universities are adopting a Bologna-inspired model is because they can use the masters section of the education to cover the things the students should have learned as undergraduates --- and thus still produce graduates with the skills that their chosen major indicates that they should have.
It's pretty damning.
The consequence is dire -- some scary example of PROFESSORS -- that is: professionals entrusted with teaching the next generation of scientists and engineers -- in the STEM fields who don't appear to understand basic thermodynamics or more specifically: entropy and the distinction between open and closed systems based on their use of thermodynamics to 'disprove' Evolution. They may be appearing to be capable professionals in every other sense and may well do 'good' research. The individuals may be appearing to do capable research in every other aspect and may be wonderful people, but their use of the 2nd law of thermodynamics as an argument against evolution is just misguided.
Andrew McIntosh -- Professor of Thermodynamics(!) and Combustion Theory at Leeds. Website.
Stuart Burgess -- Professor, Department of Engineering at Bristol. Website
In fact, the list here would presumably include mostly people of a similar persuasion. While I've seen Andrew's and Stuart's writings, I feel comfortable commenting on their opinions, but since I am not as confident about the rest of the people on the list, let's just highlight the fact that it include people from (the universities of) Sheffield, Cambridge, Liverpool and Cardiff.
Or what about this letter: http://www.bcseweb.org.uk/index.php/Main/EstelleMorris
The fact that the signatories mention their affiliations is an obvious way of trying ot use those affiliations to attach significance to their views.
Anyway, here's MC Hawking's take on it: http://www.mchawking.com/includes/lyrics/entropy_lyrics.php
22 June 2012
199. NeCTAR -Virtualisation of Australian compute resources -- first steps
So they are seeing whether they can make more efficient use of the compute resources at different institutions in Australia by creating a cloud to pool their resources. One of the potential solutions is called NeCTAR.
Getting started
Go to http://dashboard.rc.nectar.org.au/auth/login/?next=/dash/
Log in using your institutions username and password
You now have two options to deal with the key issue:
Method 1 -- generate online
Once you're in, create a keypair under Manage Compute/Access & Security and give it an easy-to-remember name
This is your private key, so protect it: don't lose it and don't expose it. You can't download it again. You delete it, it's gone.
On your computer
mv ~/Downloads/nectar.pem ~/.ssh
chmod og-rwx ~/.ssh/nectar.pem
cp nectar.pem nectar
ssh-keygen -e -f nectar >nectar.pub
ls nectar* -lah
Method 2 -- BYOK
You're using linux -- you probably have your own key already.
Go to the Manage Compute/Access & Security, Import Keypair
Paste your ~/.ssh/id_rsa.pub (or id_dsa.pub) key.
And that's the extent of the setup.
Test run
Go to Manage Compute/Images & Snapshot and select a Real Linux image (i.e. Debian)
Hit Launch.
Set up the image -- the defaults are ok, but make sure to check icmp (to be able to ping) and ssh (to be able to log in).
Generating and loading the image takes about 10-20 seconds -- about the duration of a Victorian earthquake.
Now your image is up an running. To check that all is well
To be able to log in via ssh you need to know what username to use -- it's (probably) image specific.
To find out, click on the image name (here: testing4)
Hit the 'Log' tab
And look for the username which is created in addition to root
Once you're in you'll be greeted by:
When you're done, don't forget to log out and terminate your image. If you leave it running it will count towards your resource allocations.
Notes:
1. you'll run into trouble with the key fingerprints eventually as the IP addresses and key fingerprints won't be matching. Either you'll be doing a lot of editing of you ~/.ssh/known_hosts file or you have to relax your security setttings.
2. Yes, you can log in as root as well. The default user does not have sudo powers.
3. It takes about 60 seconds after the launch of the image before the openssh server is up and accepting connections. Think more desktop speeds than laptop+SSD speeds.
4. For actual production stuff you can crank up the image requirements:
5. Also, I think the real value of using virtual machine is that you can load a vanilla setup and customize it, then saving it by making a snapshot:
A first couple of actions might be to add a new user, and edit /etc/sudoers.
Troubleshooting ssh:
If you're having problems logging in using your key, use the ssh -v switch as shown above and parse the output
Unsuccessful:
A successful authentication should contain
If you are sure that you're using the right key (e.g. using -i), then make sure that you're using the right username -- to find out how to find it, look above.
Getting started
Go to http://dashboard.rc.nectar.org.au/auth/login/?next=/dash/
Log in using your institutions username and password
You now have two options to deal with the key issue:
Method 1 -- generate online
Once you're in, create a keypair under Manage Compute/Access & Security and give it an easy-to-remember name
This is your private key, so protect it: don't lose it and don't expose it. You can't download it again. You delete it, it's gone.
On your computer
mv ~/Downloads/nectar.pem ~/.ssh
chmod og-rwx ~/.ssh/nectar.pem
cp nectar.pem nectar
ssh-keygen -e -f nectar >nectar.pub
ls nectar* -lah
-rw------- 1 me me 887 Jun 22 11:31 nectar
-rw------- 1 me me 887 Jun 22 11:28 nectar.pem
-rw-r--r-- 1 me me 335 Jun 22 11:31 nectar.pub
To use the key do
ssh -i nectar user@server
Method 2 -- BYOK
You're using linux -- you probably have your own key already.
Go to the Manage Compute/Access & Security, Import Keypair
Paste your ~/.ssh/id_rsa.pub (or id_dsa.pub) key.
And that's the extent of the setup.
Test run
Go to Manage Compute/Images & Snapshot and select a Real Linux image (i.e. Debian)
 |
| Select image |
 |
| Set up -- don't forget to check SSH to be able to log on. If you want to be able to ping, check icmp as well. |
 |
| Loading |
 |
| Running |
ping -c 3 115.146.92.154
PING 115.146.92.154 (115.146.92.154) 56(84) bytes of data. 64 bytes from 115.146.92.154: icmp_req=1 ttl=54 time=1.70 ms 64 bytes from 115.146.92.154: icmp_req=2 ttl=54 time=1.66 ms 64 bytes from 115.146.92.154: icmp_req=3 ttl=54 time=1.73 ms --- 115.146.92.154 ping statistics --- 3 packets transmitted, 3 received, 0% packet loss, time 2002ms rtt min/avg/max/mdev = 1.660/1.698/1.733/0.029 ms
To be able to log in via ssh you need to know what username to use -- it's (probably) image specific.
To find out, click on the image name (here: testing4)
 |
| Click me |
 |
| Select 'log' |
 |
| Look for the username -- here it's debian |
ssh -v -i ~/.ssh/tmp/nectar debian@115.146.92.154 The authenticity of host '115.146.92.154 (115.146.92.154)' can't be established. RSA key fingerprint is 81:a8:a7:0f:a9:68:a0:08:f1:60:45:e3:57:2e:4c:4c. Are you sure you want to continue connecting (yes/no)? yes Warning: Permanently added '115.146.92.154' (RSA) to the list of known hosts.
Once you're in you'll be greeted by:
Linux unnamed-virtual-machine 2.6.32-5-amd64 #1 SMP Thu Mar 22 17:26:33 UTC 2012 x86_64 The programs included with the Debian GNU/Linux system are free software; the exact distribution terms for each program are described in the individual files in /usr/share/doc/*/copyright. Debian GNU/Linux comes with ABSOLUTELY NO WARRANTY, to the extent permitted by applicable law. debian@i-00001637:~$
debian@i-00001637:~$ df -h Filesystem Size Used Avail Use% Mounted on /dev/vda1 9.9G 768M 8.6G 9% / tmpfs 2.0G 0 2.0G 0% /lib/init/rw udev 2.0G 112K 2.0G 1% /dev tmpfs 2.0G 0 2.0G 0% /dev/shm debian@i-00001637:~$ uname -a Linux i-00001637 2.6.32-5-amd64 #1 SMP Thu Mar 22 17:26:33 UTC 2012 x86_64 GNU/Linux debian@i-00001637:~$ groups debian cdrom floppy audio dip video plugdev debian@i-00001637:~$ cat /etc/group|grep debian cdrom:x:24:debian floppy:x:25:debian audio:x:29:debian dip:x:30:debian video:x:44:debian plugdev:x:46:debian debian:x:1000:
When you're done, don't forget to log out and terminate your image. If you leave it running it will count towards your resource allocations.
 |
| Terminating |
Notes:
1. you'll run into trouble with the key fingerprints eventually as the IP addresses and key fingerprints won't be matching. Either you'll be doing a lot of editing of you ~/.ssh/known_hosts file or you have to relax your security setttings.
2. Yes, you can log in as root as well. The default user does not have sudo powers.
3. It takes about 60 seconds after the launch of the image before the openssh server is up and accepting connections. Think more desktop speeds than laptop+SSD speeds.
4. For actual production stuff you can crank up the image requirements:
 |
| 16 cores and 65 GB? Why, thank you! |
 |
| Snapshot saves |
Troubleshooting ssh:
If you're having problems logging in using your key, use the ssh -v switch as shown above and parse the output
Unsuccessful:
debug1: Authentications that can continue: publickey,password debug1: Next authentication method: publickey debug1: Offering RSA public key: /home/me/.ssh/id_rsa debug1: Authentications that can continue: publickey,password debug1: Trying private key: /home/me/.ssh/id_dsa debug1: Trying private key: /home/me/.ssh/id_ecdsa debug1: Next authentication method: password
A successful authentication should contain
debug1: Offering RSA public key: /home/me/.ssh/id_rsa debug1: Server accepts key: pkalg ssh-rsa blen 279 debug1: read PEM private key done: type RSA debug1: Authentication succeeded (publickey).
21 June 2012
197. Post-mortem of the Moe Quake
The news keep on reporting about a single injured person who was unlucky enough to be standing on a ladder close to the epicentre, but beyond that it seems like no-one else suffered any injuries serious enough to warrant medical attention.
Property damages are a different story though, but the news are having a field day with it, so no point in me repeating what they are saying. My only comment is that the Gippsland/La Trobe Valley area has been hit hard lately, first by floods, storms and now an earthquake -- in addition to recent job losses and uncertainties.
Anyway, science.
www.ga.gov.au has a nice page with technical information about the earthquake:
http://www.ga.gov.au/earthquakes/getQuakeDetails.do?quakeId=3226344
There are several seismograms available from different stations around the country (am I the only chemist who looks at them wanting to apply a FFT?)
The shape varies with the distance from the earthquake, which I guess tallies with different types of waves travelling at different speeds.
For those of us who are reasonably new to this area, the USGS has a historical earthquake map over Melbourne and the Gippsland/La Trobe valley area.
Here's seismicity in Australia as a whole, and it shows that SE Victoria is no stranger to phenomenon:
http://www.quakes.uq.edu.au/html/quake_info/OZ_QLD_info.html#6
Big earthquakes are a different matter though: http://earthquake.usgs.gov/earthquakes/world/australia/seismicity.php
Only a handful of earthquakes show up on this map, and they are in WA and NT.
Here's a map with the number of large earthquakes per year (5 and above) -- and Melbourne is by no means the worst hit by the Top 5 cities in Australia
http://earthquake.usgs.gov/earthquakes/world/australia/density.php
Finally, here's a map with the 'earthquake hazard' estimates for different regions of Australia:
http://earthquake.usgs.gov/earthquakes/world/australia/gshap.php
It seems like SE Tassie is the safest, inhabited area. SW WA is the least safe one, but is still nothing compared to PNG and Indonesia and other countries on plate boundaries.
Here's a full paper on seismic hazards in Australia, which contains a nice map with past earthquakes indicated on it: http://www.sciencedirect.com/science/article/pii/S0040195104002185
Property damages are a different story though, but the news are having a field day with it, so no point in me repeating what they are saying. My only comment is that the Gippsland/La Trobe Valley area has been hit hard lately, first by floods, storms and now an earthquake -- in addition to recent job losses and uncertainties.
Anyway, science.
www.ga.gov.au has a nice page with technical information about the earthquake:
http://www.ga.gov.au/earthquakes/getQuakeDetails.do?quakeId=3226344
There are several seismograms available from different stations around the country (am I the only chemist who looks at them wanting to apply a FFT?)
 |
| Victoria |
 |
| ACT |
 |
| Northern QLD |
The shape varies with the distance from the earthquake, which I guess tallies with different types of waves travelling at different speeds.
For those of us who are reasonably new to this area, the USGS has a historical earthquake map over Melbourne and the Gippsland/La Trobe valley area.
Here's seismicity in Australia as a whole, and it shows that SE Victoria is no stranger to phenomenon:
http://www.quakes.uq.edu.au/html/quake_info/OZ_QLD_info.html#6
Big earthquakes are a different matter though: http://earthquake.usgs.gov/earthquakes/world/australia/seismicity.php
Only a handful of earthquakes show up on this map, and they are in WA and NT.
Here's a map with the number of large earthquakes per year (5 and above) -- and Melbourne is by no means the worst hit by the Top 5 cities in Australia
http://earthquake.usgs.gov/earthquakes/world/australia/density.php
Finally, here's a map with the 'earthquake hazard' estimates for different regions of Australia:
http://earthquake.usgs.gov/earthquakes/world/australia/gshap.php
It seems like SE Tassie is the safest, inhabited area. SW WA is the least safe one, but is still nothing compared to PNG and Indonesia and other countries on plate boundaries.
Here's a full paper on seismic hazards in Australia, which contains a nice map with past earthquakes indicated on it: http://www.sciencedirect.com/science/article/pii/S0040195104002185
 |
| I'm loading the picture from the publisher's website which is probably the lesser of two evils. |
28 May 2012
167. ECCE/Nwchem on An Australian University computational cluster using qsub with g09/nwchem
EDIT:
I've just learned the First Rule of Remote Computing:
always start by checking the number of concurrent processes you're allowed on the head node, or you can lock yourself out faster that you can say "IT support'.
do
ulimit -u
If it's anywhere under 1000, then you need to be careful.
Default ulimit on ROCKS: 73728
Default ulimit on Debian/Wheezy: 63431
Ulimit on the Oz uni cluster: 32
ECCE launches FIVE processes per job.
Each pipe you add to a command launches another proc. Logging in launches a proc -- if you've reached your quota, you can't log in until a processes finishes.
cat test.text|sed 's/\,/\t/g'|gawk '{print $2,$3,$4}'
yields three processes -- ten percent of my entire quota.
NOTE:
Running something on a cluster where you have limited access is very different from a cluster you're managing yourself. Apart from knowing the physical layout, you normally have sudo powers on a local cluster.
On potential issue is excessive disk usage -- both in terms of storage space and in terms of raw I/O (writing to an nfs mounted disk is not efficient anyway)
So in order to cut down on that:
1. Define a scratch directory using e.g. (use the correct path)
scratch_dir /scratch
The point being that /scratch is a local directory on the execution node
2. Make sure that you specify
I accidentally ended up storing 52 GB of aoints files from a single job. It may have been what locked me out of the submit node for three hours...
A good way to check your disk-usage is
ls -d * |xargs du -hs
Now, continue reading:
Setting everything up the first time:
First figure out where the mpi libs are:
qsub.tests:
Next, look at ls /opt/sw/nwchem-6.1/data -- if there's a default.nwchemrc file, then
ln -s /opt/sw/nwchem-6.1/data/default.nwchemrc ~/.nwchemrc
If not, create ~/.nwchemrc with the locations of the different basis sets, amber files and plane-wave sets listed as follows:
Using nwchem:
A simple qsub file would be:
with test.nw being the actual nwchem input file which is present in your cwd (current working directory).
Using nwchem with ecce:
This is the proper way of using nwchem. If you haven't already, look here: http://verahill.blogspot.com.au/2012/05/setting-up-ecce-with-qsub-on-australian.html
Then edit your ecce-6.3/apps/siteconfig/CONFIG.msgln4 file:
Since I'm having problems with the low ulimit, I wrote a small bash script which I've set to run every ten minutes as a cronjob. Of course, if you've used up your 32 procs you can't run the script...also, instead of piping stuff right and left (each pipe creates another fork/proc) I've written it so it dumps stuff to disk. That way you have a list over procs in case you need to kill something manually:
The script: ~/clean_ps.sh
and the cron job is set up using
crontab -e
*/10 * * * * sh ~/clean_ps.sh>> ~/.cronout
Obviously this kills any job monitoring from the point of view of ecce. However, it keeps you from being locked out. You can manually check the job status using qstat -u ${USER}, then reconnect when a job is ready. Not that convenient, but liveable.
I've just learned the First Rule of Remote Computing:
always start by checking the number of concurrent processes you're allowed on the head node, or you can lock yourself out faster that you can say "IT support'.
do
ulimit -u
If it's anywhere under 1000, then you need to be careful.
Default ulimit on ROCKS: 73728
Default ulimit on Debian/Wheezy: 63431
Ulimit on the Oz uni cluster: 32
ECCE launches FIVE processes per job.
Each pipe you add to a command launches another proc. Logging in launches a proc -- if you've reached your quota, you can't log in until a processes finishes.
cat test.text|sed 's/\,/\t/g'|gawk '{print $2,$3,$4}'
yields three processes -- ten percent of my entire quota.
NOTE:
Running something on a cluster where you have limited access is very different from a cluster you're managing yourself. Apart from knowing the physical layout, you normally have sudo powers on a local cluster.
On potential issue is excessive disk usage -- both in terms of storage space and in terms of raw I/O (writing to an nfs mounted disk is not efficient anyway)
So in order to cut down on that:
1. Define a scratch directory using e.g. (use the correct path)
scratch_dir /scratch
The point being that /scratch is a local directory on the execution node
2. Make sure that you specify
dftor even
direct
..
end
dftto do as little disk caching as possible.
noio
...
end
I accidentally ended up storing 52 GB of aoints files from a single job. It may have been what locked me out of the submit node for three hours...
A good way to check your disk-usage is
ls -d * |xargs du -hs
Now, continue reading:
Setting everything up the first time:
First figure out where the mpi libs are:
qsub.tests:
#!/bin/sh
#$ -S /bin/sh
#$ -cwd
#$ -l h_rt=00:14:00
#$ -l h_vmem=4G
#$ -j y
locate libmpi.so
Assuming that the location is /usr/lib/openmpi/1.3.2-gcc/lib/, put
export LD_LIBRARY_PATH=/usr/lib/openmpi/1.3.2-gcc/lib/
in your ~/.bashrc
Next, look at ls /opt/sw/nwchem-6.1/data -- if there's a default.nwchemrc file, then
ln -s /opt/sw/nwchem-6.1/data/default.nwchemrc ~/.nwchemrc
If not, create ~/.nwchemrc with the locations of the different basis sets, amber files and plane-wave sets listed as follows:
nwchem_basis_library /opt/sw/nwchem-6.1/data/libraries/
nwchem_nwpw_library /opt/sw/nwchem-6.1/data/libraryps/
ffield amber
amber_1 /opt/sw/nwchem-6.1/data/amber_s/
amber_2 /opt/sw/nwchem-6.1/data/amber_q/
amber_3 /opt/sw/nwchem-6.1/data/amber_x/
amber_4 /opt/sw/nwchem-6.1/data/amber_u/
spce /opt/sw/nwchem-6.1/data/solvents/spce.rst
charmm_s /opt/sw/nwchem-6.1/data/charmm_s/
charmm_x /opt/sw/nwchem-6.1/data/charmm_x/
Using nwchem:
A simple qsub file would be:
#!/bin/sh
#$ -S /bin/sh
#$ -cwd
#$ -l h_rt=00:14:00
#$ -l h_vmem=4G
#$ -j y
#$ -pe orte 4
module load nwchem/6.1
time mpirun -n 4 nwchem test.nw > nwchem.out
with test.nw being the actual nwchem input file which is present in your cwd (current working directory).
Using nwchem with ecce:
This is the proper way of using nwchem. If you haven't already, look here: http://verahill.blogspot.com.au/2012/05/setting-up-ecce-with-qsub-on-australian.html
Then edit your ecce-6.3/apps/siteconfig/CONFIG.msgln4 file:
NWChem: /opt/sw/nwchem-6.1/bin/nwchem
Gaussian-03: /usr/local/bin/G09
perlPath: /usr/bin/perl
qmgrPath: /usr/bin/qsub
SGE {
#$ -S /bin/csh
#$ -cwd
#$ -l h_rt=$wallTime
#$ -l h_vmem=4G
#$ -j y
}
NWChemFilesToDelete{ core *.aoints.* }
NWChemEnvironment{
LD_LIBRARY_PATH /usr/lib/openmpi/1.3.2-gcc/lib/
}
NWChemCommand {
#$ -pe mpi_smp4 4
module load nwchem/6.1
mpirun -n $totalprocs $nwchem $infile > $outfile
}
Gaussian-03Command {
#$ -pe g03_smp4 4
module load gaussian/g09
time G09< $infile > $outfile }
Gaussian-03FilesToDelete{ core *.rwf }
Wrapup{
find /scratch/* -name "*" -user $USER |xargs -I {} rm {} -rf
}
And you should be good to go. IMPORTANT: don't copy the settings blindly -- what works at your uni might be different from what works at my uni. But use the above as an inspiration and validation of your thought process. The most important thing to look out for in terms of performance is probably your -pe switch.
Since I'm having problems with the low ulimit, I wrote a small bash script which I've set to run every ten minutes as a cronjob. Of course, if you've used up your 32 procs you can't run the script...also, instead of piping stuff right and left (each pipe creates another fork/proc) I've written it so it dumps stuff to disk. That way you have a list over procs in case you need to kill something manually:
The script: ~/clean_ps.sh
date
ps ux>~/.job.list
ps ux|gawk 'END {print NR}'
cat ~/.job.list|grep "\-sh \-i">~/.job2.list
cat ~/.job2.list|gawk '{print$2}'>~/.job3.list
cat ~/.job3.list|xargs -I {} kill -15 {}
cat ~/.job.list|grep "echo">~/.job4.list
cat ~/.job4.list|gawk '{print$2}'>~/.job5.list
cat ~/.job5.list|xargs -I {} kill -15 {}
cat ~/.job.list|grep "notty">~/.job6.list
cat ~/.job6.list|gawk '{print$2}'>~/.job7.list
cat ~/.job7.list|xargs -I {} kill -15 {}
cat ~/.job.list|grep "perl">~/.job8.list
cat ~/.job8.list|gawk '{print$2}'>~/.job9.list
cat ~/.job9.list|xargs -I {} kill -15 {}
qstat -u ${USER}
ps ux |gawk 'END {print NR}'
echo "***"
and the cron job is set up using
crontab -e
*/10 * * * * sh ~/clean_ps.sh>> ~/.cronout
Obviously this kills any job monitoring from the point of view of ecce. However, it keeps you from being locked out. You can manually check the job status using qstat -u ${USER}, then reconnect when a job is ready. Not that convenient, but liveable.
15 February 2012
65. LaTeX for 2013 ARC proposals
If you're in the limbo where you're not quite expert enough with LaTeX, but have happily said goodbye to MS Office, and you're an Australia scientist who annually has to fight the endless bureaucracy of the Australian Research Council (seriously -- 10 page limit about the science, followed by 70 pages about personal qualities?) you may find that you're in for a lot of grief.
If you collaborate with other people you will have to settle for a file format everyone can deal with ('everyone' except linux users), which is likely to be Word run in a virtual environment. If you're writing by yourself or together with more computer-savvy colleagues you will get away with using LaTeX for a fair number of the sections, while using gEdit for others.
I'm presuming that you will create a separate .tex file for each section.
The rules (http://www.arc.gov.au/ncgp/dp/dp_instructions.htm) says that you need
Also, reading between the lines, you shouldn't/needn't include page numbers since this will be generated by the RMS.
with \pagestyle{plain} just after \begin{document} may do the trick.
Black type and single column are defaults in LaTeX, so need no action. Everything below goes into the preamble.
For A4 and 12 point font, set that in the documentclass
Part E, while a pain to write, uses regular 12 pt Times New Roman. If you don't want section numbering, use {secnumbdepth}{0}. If you want your sections to look like this
If you collaborate with other people you will have to settle for a file format everyone can deal with ('everyone' except linux users), which is likely to be Word run in a virtual environment. If you're writing by yourself or together with more computer-savvy colleagues you will get away with using LaTeX for a fair number of the sections, while using gEdit for others.
I'm presuming that you will create a separate .tex file for each section.
The rules (http://www.arc.gov.au/ncgp/dp/dp_instructions.htm) says that you need
- Black type
- Single column
- A4
- 0.5 cm margin on all sides (top, bottom, right, left)
- A 'highly legible font', preferably 12 pt Times New Roman
Also, reading between the lines, you shouldn't/needn't include page numbers since this will be generated by the RMS.
\usepackage{nopageno}
with \pagestyle{plain} just after \begin{document} may do the trick.
Black type and single column are defaults in LaTeX, so need no action. Everything below goes into the preamble.
For A4 and 12 point font, set that in the documentclass
\documentclass[12pt,a4paper]{article}\usepackage[latin1]{inputenc}\usepackage{amsmath}\usepackage{amsfonts}\usepackage{amssymb}\usepackage{mhchem}
For 0.5 cm margins on all sides
Part C, section C1 Here you have several sections, such as AIMS AND BACKGROUND, RESEARCH PROJECT etc. To make sure that these DO NOT get numbered, use
\seccounter{secnumbdepth}{0}
Since the references are a separate document, set\documentclass[10pt,a4paper]{article}Part D is filled out online.
Part E, while a pain to write, uses regular 12 pt Times New Roman. If you don't want section numbering, use {secnumbdepth}{0}. If you want your sections to look like this
E1. Justification....
E2. Details of non-ARC contributions
Use
\seccounter{secnumbdepth}{1}\def\thesection{E\arabic{section}.}
As usual, this goes into the preamble, not the body. You should -- ideally -- never have to change the way you write. All changes should go into the documentclass style file or preamble.
Part F. Oh how we curse you.
F13.1 should be easy enough -- you can even skip sections and just use \title{F13.1. Justification} together with a \maketitle right after \begin{document}. Or you can use
\def\thesection{F13.\arabic{section.}
F14.2 offers potential for fun:
Part F. Oh how we curse you.
F13.1 should be easy enough -- you can even skip sections and just use \title{F13.1. Justification} together with a \maketitle right after \begin{document}. Or you can use
\def\thesection{F13.\arabic{section.}
F14.2 offers potential for fun:
\setcounter{secnumdepth}{4}
\def\thesection{F14.2.\arabic{section}.}
\def\thesubsection{\alph{subsection})}
\title{F14.2 Recent significant publications (since 2007)}
This way if you put the following in the body:
\maketitle
\section{Research publications}
\subsection{Scholarly books}
\subsection{Scholarly book chapters}
you get
F14.2 Recent significant publications (since 2007) F14.2.1. Research publications a) Scholarly books b) Scholarly book chapters
For your reference list you may use
25.* Blow, Joe; Doe, John; \textbf{Hill, Vera}. \emph{One page derivation of the GUT}, Phys. Rev., \textbf{2012}, \emph{108(3)}, 234-235 \\For F14.3 use\def\thesection{F\arabic{section}.} \def\thesubsection{14.\arabic{subsection}}without \maketitle but with\subsection{Ten career-best publications}in the bodyG1 is difficult in latex if you want to do it properly. I gave up and used libreoffice for this one since I couldn't figure out howPut this in the preamble\usepackage{rotating} %rotate text in narrow tables...\usepackage{array} % for vertical centering in tables...hacky...and this in the body:\section{Research support for Vera Hill}\begin{table}[h]\begin{tabular}{b{4cm} p{0.5cm} p{0.5cm} l c c c c c}\textbf{ Description} & \begin{sideways}Same Research area \end{sideways} & \begin{sideways}Support type \end{sideways}& ARC Project ID & 2010 (k\$) & 2011 (k\$) & 2012 (k\$) & 2013 (k\$) & 2014 (k\$) \\\hline"Unifying general relativity with quantum mechanics" & N & C & DP100202921 & 120 & 190 & 180 & 170 & 120 \\\hline\end{tabular}\end{table}It doesn't look quite right. You may also want to try e.g. m{0.5cm} for the rotated header captions The remaining sections should be fairly straightforward, formatting-wise. Ultimately, the ARC application process is one of the most ridiculous one I've had the misfortune to be subjected to.
Subscribe to:
Posts (Atom)