SOLID2000 is a simulation program for systems that have symmetry in three dimensions, such as crystals. The code is written mostly in FORTRAN77 and FORTRAN95.
My project is to parallelize the calculation of the band structure using MPI, which is basically the energy of the electrons inside the solid. Such solutions are impossible to find analytically for the real word systems, that’s why we have to use numerical methods to sample part of our solution.
The finest (dense) our sampling is, the better our plot will be, and the more calculation time is required. For every sample we make, we have to solve an NxN matrix, i.e to find its eigenvalues. Those eigenvalues are the solution we desire, i.e the energy of the electrons, and the matrix is called the Hamiltonian matrix which describes the whole energy of our system, potential energy + the kinetic energy. Normally the size of the matrix is infinite so it’s up to “cut” the matrix.
The time required to calculate the eigenvalues scales as N3 and therefore we want our matrix to be as small as possible but not too small because the accuracy of our results will be very low. Our strategy to parallelize the band structure is to assign every sampling to a procedure up until all the sample points are done.
This will be done by a message that the master will send to all the slaves to signal that the band structure calculation is about to start and each procedure will calculate its sample points. The work will be distributed equally throughout the procedures in order to have the maximum efficiency. At the end of the calculation, the slaves must send their samples to the master so the master can write the band structure in a file. The next picture shows an example of a band structure of a carbon nanotube which was generated using this code.
Band structure of a carbon nanotube where y-axis is the Energy and x-axis is the k-point (or the sample point)
James and I went on a tour to visit the machines we’re working with, Salomon and Anselm supercomputers. We were guided by the IT4I institute director Branislav Jansík, and it was soon apparent he is acquainted with just about every detail of the inner workings of the centre and its supercomputers.
Salomon Supercomputer
The tour started in a meeting room, behind thick glass through which we could see the computers running. And you can believe that glass was thick, as we could only faintly hear the noise coming from the other side. As we later learned, supercomputers are so loud you will have trouble holding a conversation next to it without it looking like a game of telephone. If that’s not enough, the noise produced by all the components that are in a centre like this can be very much heard outside the actual building.
Soon after, we were led into the spacious room where the supercomputers stood. Before entering, however, we were warned that the oxygen level in the room was around 15%, compared to the normal ~21%, which can cause dizziness or even passing out. There’s a good reason for this, however – oxygen at this level prevents fire from spreading in case of an emergency. As a matter of fact, if you were to a light a candle, it would just be smothered soon after by the lack of oxygen. Fortunately, we haven’t experienced any problems and proceeded to view the components of a supercomputer.
Branislav Jansík showing us a rack with storage.
The components of a supercomputer are neatly divided into storage, networking and processing units. Most of the racks you see the in the picture are processing units, which do the heavy lifting of high performance computing. A few others are dedicated to connecting these nodes in such a way that they can communicate efficiently, which usually involves multidimensional geometry in order to make sure no node is too far away from another topologically. And of course, there are storage nodes which hold all the data. There is actually so much data, more than a petabyte, in this case, it’s impractical to back it all up, so usually only one small portion of it is.
The computer room is far from being the only large part of a supercomputing centre, as the next stop was the infrastructure required to sustain this beast.
The heating system of IT4I.
After leaving the computer room, we headed towards the roof and stopped at a section which deals with drawing away all the generated heat. And there is heat being generated alright! There is so much of it, they use it to keep the building warm, even during the winter, with no additional source! Moreover, only 10% of the generated heat is sufficient for this purpose.
The electricity bill is no joke, however, and if you’re interested, it’s in the order of tens of thousands of euros a month. Moving on, we went to the roof.
Roof of IT4I.
Arriving atop the building, we were mostly surrounded by fences around the cooling towers and chillers, and we happened to find a plant thriving without a care in the world among all the metal around it.
Anyway, this is where the tour ended, wherefrom Branislav showed us a secret shortcut door that led straight to our office, in which we could, again, tap into Salomon’s power solely through a console interface. It’s incredible to see what actually goes into the machinery that we have been using for weeks now through a very convenient and straightforward terminal on our laptops.
Life is fantastic in Ljubljana. Working in the surroundings of wonderful mountains makes me feel like a hero of a fairy tale. We’ve changed the month in the calendar but I still can’t believe how fast time passes.
During the PRACE Summer of High Performance Computing programme I am working on big data using RHadoop. The data set is a set of 80 million ionizing radiation level measurements in the world. To explain everything clearly, I have to do a little introduction.
Beautiful evening near Ljubljana castle.
Theory
First of all we have to see the difference between types of radiation. We can divide them into two basic types:
– Non ionizing radiation – electromagnetic radiation that does not carry enough energy per quantum to ionize atoms or molecules.
– Ionizing radiation – carries enough energy to liberate electrons from atoms or molecules, thereby ionizing them.
Ionizing radioactivity can be both our ally and enemy – it all depends on the size of the dose. It comes from the cosmos, industry, mines and even from another person. It is also very useful tool in medicine. There are many treatments based on ionizing radiation such as x-rays or computed tomography.
In most types of examinations connected with it, the value is so small that it requires hundreds or thousands of repetitions to threaten our health. The most important thing for us to know, is how much radiation can we receive.
There are many units describing radiation level (Rad,Gray,Curie,Sievert) – I will use Sieverts ( 1Sv = 1J/1Kg).
Exposure to 100 mSv a year is the lowest level at which any increase, is clearly evident a cancer risk. For example, a chest x-ray provides only 20 µSv, while a mammogram procedure provides 3.0 mSv.
People have contact with radiation every day in many fields. Yearly dose for natural potassium in the body is 390 µSv and even eating one banana is connected with a dose of 0.1 µSv.
Dependence of wavelength on energy
HPC adventure
We’ve installed RHadoop software on the HPC infrastructure which allows to manage really huge data. In my case, the text file has 10 GB of data, but RHadoop is ready to manage much bigger files. In the data set there are radiation measurements provided by a volunteer science project – ‘SAFECAST’.
What I intend to do with the data set is to use clustering algorithms to detect most frequent regions with high level of radiations. Different clustering algorithms process the data from many sides – for example we can check the highest (lowest) radiations levels only in January.
We use RHadoop which allows for concurrent processing. Primarily, a master-node divides the data into smaller, independent chunks. Then each worker node works on one of them. It uses a Map function which performs filtering and sorting. The next step is Shuffle – worker nodes redistribute data based on the output keys. The output goes to the Reduce function which processes each group of output data, per key, in parallel.
So What’s next ?
I am planning to use machine learning ideas to support clustering algorithms and to make an independent way of research. To be honest – I can’t wait to compare the results !
Some of the first results of Hadoop processing. European places and CPM (Counts Per Minute – another radiation unit) values in logarithmic scale.
It’s already august. This summer is going by really quickly. We have already worked a lot, but there are still many things to do till the end of of this month!
About myself and about Slovenia, almost nothing has changed since the last post. Ljubljana is still wonderful and Slovenia is still fantastic. The more places I visit, the more I like it.
Slovenia is full of beautiful caves!
The weather is not so hot and everything is really green. Also, the people in the faculty here are very friendly, and it’s always nice to work with them.
About my project, we have already created not only a Plugin for Paraview, but also a tool that makes very easy to create those plugins! We have also managed to do everything in Python, which has some advantages in relation to the “traditional” C++ way:
Comparative scheme between the C++ way” and the “Python Way” to create a plugin for ParaView.
With this Plugin, we can easily take data from ITER’s so called IMAS database and play with it. For example, we can actually extract the information about some quantities such as temperature or magnetic strength and plot them in the grid, or study the evolution of the dataset with time.
A simple animation that shows the evolution of Magnetic field in the ITER’s vacuum vessel during a shot, loaded with our Plugin
We will continue upgrading the plugin and adding more features to it in the following days. Making things easier is not an easy task!
If you want more information, I’ve been recently interviewed about this project in a funny TV show that actually does not exist. The following video contains the full interview. Enjoy!
I’m well underway into the PRACE Summer of HPC program and my project is going well. I am working with Dr. Nick Johnson (link) on the NEXTGENIO project (link). Here is an introduction to my project.
HPC systems are expensive machines, and running jobs on them uses up precious resources and funding. The study of how jobs get allocated to different nodes could maximise the usage of these resources. The goal of my project is to explore how we can utilise our resources in an optimal way by testing different job scheduling algorithms.
When a person wants to do some work on a HPC system they are allocated some nodes on which they can run their code for whatever their purposes may be, a person may be given 8 nodes typically. To give some scale about the capacity of the HPC system called `Archer’ at EPCC, it has 4920 nodes on which users can run simulations, calculations or whatever their desire may be. Each node has 24 processors, so that’s over 100,000 processors. There are lots of jobs running on Archer all the time and also lots of requests for nodes to be allocated. So it is of vital importance to study how these nodes are allocated to different users so that we minimise the number of inactive nodes during a time slot and maximise the number of jobs performed.
Diagram of jobs running on a HPC system. Some nodes are left vacant while other perform multiple jobs. Note how the diagram highlights the time spent Reading and Writing. This can be reduced by using NNVRAM if the same nodes are used again. This is one of the topics of investigation during this project.
Different users may request different numbers of nodes (depending on the size of their project) and different lengths of time for how long they will need them. Suppose Alice requests 100 nodes for 8 hours and then Bob requests 10 nodes for 2 hours. If we were to operate on a first-come first serve basis (a very näive approach), then both Alice and Bob must wait for the 100 nodes to become free before Alice can run her jobs and then Bob can run his. However, suppose 20 nodes were free the whole time and Alice must wait more than 2 hours before 100 nodes are free to begin her job. A more efficient algorithm would allow Bob to skip Alice in the queue and let him perform his job on these nodes that she can’t use. He will not be delaying the time at which she begins her jobs as she will still be waiting for the other 80 of the 100 nodes to become free when Bob is finished. This is a very simple two job example to demonstrate how we can have 2 different job scheduling algorithms. Clearly the second algorithm will reduce the overall time to perform the two jobs in this situation.
This is a simple case, but it serves as an example to see the importance in job scheduling on a HPC system, otherwise queues for jobs could be days long. If you’d like to learn more, here is a paper I’ve been reading about it (link).
Free time
I’m a very keen golfer, having represented my university for 4 years, and Captaining the team. Being in Scotland, the glorified home of golf, it is pretty special, especially at this time of year when the Scottish and British Opens are on back to back weeks and in close proximity to Edinburgh. So it was on my agenda to make sure I went to one of golf’s 4 majors, the Open at Carnoustie golf club. My dad flew in and we made a trip of it. Seeing Tiger Woods return to some of his old form, McIlroy hit Driver-Sand Wedge to a par 5, Jason Day hit a wedge stone dead to a foot and the eventual winner Francseco Molinari were the standout moments. A great day, to get the head cleared after a tough week of coding.
Me at the British Open
The Iconic 18th hole, where tournaments are won and lost; just ask Molinari and Van de Velde.
Hi everyone! Welcome back! It’s has been three weeks since we arrived in Nicosia. As an introduction to Cyprus culture, our site coordinator (shout-out to Stelios!) took us out for a meze the night we landed. I’ve never seen so many dishes for just 3 people! On top of that, everything tasted amazing, specially the halloumi, a mixture of goat and sheep cheese which is usually served grilled.
On the next day, we went around the old part of Nicosia (inside the Venetian walls), and it was quite difficult to find our way, since it looked like a labyrinth. But now we know it like the palm of our hand. Also, who would have known that the usage of Greek letters in physics problems would turn out to be really useful (even though we don’t understand Greek).
Fig. 1: Walking route marker found around the old part of Nicosia. Notice the shape of the Venetian Walls.
Back to work. As I mentioned in the introductory post, here I’m going to explain what my project is about and what I’ll have to do. But on second thought I’m going to give you guys a little bit of background.
First of all, we need to know a bit of physics (don’t be scared). In school, we all learned that there are 4 fundamental forces that govern how particles interact with each other: gravity, electromagnetism, strong and weak. The one that I’m interested in is the strong interaction, and the theory behind it is called Quantum Chromodynamics (or QCD).
With only two ingredients, quarks and gluons, QCD tries to explain a wide range of phenomena: from the collisions taking place at LHC (very high energy, around the TeV scale) to how the protons and nucleons are squeezed together to form the nucleus of the atoms (very small energy, around the MeV scale). The problem is that, while for the first case we can manage to understand the process using pen and paper, for the second one it is very difficult, since all our mathematical tools break down at this energy regime.
This obstacle has not stopped physicists from making predictions at this scale, even though they can no longer be from first principles and depend on experimental data. If we want to use QCD directly and compute (for example) how the mass of the proton emerges from the interaction between the quarks and the gluons, the only way to do it is by using computers. Or, to be more precise, supercomputers.
To do that, we need to program QCD on a computer. This has been known for a couple of decades: lattice QCD (LQCD). It has the word lattice in it because what we do is discretize the space-time in a 4-dimensional grid, placing the quarks on the nodes of this lattice and the gluons on the links. Then, if we want to simulate the proton, we put 3 quarks (2 up and 1 down quarks) on the lattice, let them evolve, and see if at the end we get a proton or not. But why is it so computationally demanding? We have to evaluate a path integral, or in other words, we have to compute every possible path that a quark can take from one point of the lattice to each and every other. And then repeat this computation several times to get enough statistics.
As an example, in Fig. 2, the masses for different baryons (particles made up of 3 quarks, like the proton) have been calculated by different collaborations, and then compared with experimental data:
Fig. 2: The colored points are the masses of the baryons computed using LQCD, and the black line (with gray band) is the experimental value [1].
As you can see, LQCD gives a result which is in perfect agreement with the experimental data. But this type of simulations couldn’t be done in previous years. In the beginning, LQCD had to use a higher value for the masses of the quarks than the physical ones, since it is easier and faster to perform the simulations. To get to the physical point, new algorithms had to be developed.
One of these is the multigrid solver. To know what it is you’ll have to wait for the next post, where I’ll also introduce my project. See you then!
So I’m over here in Edinburgh working on a project in the field of “High Performance Computing” (HPC), but what does that mean? I didn’t know what that was when I was first introduced to it a year ago, so I understand if it’s raising a few eyebrows. Maybe what’s running through your head is the scene from House of Cards where Lucas Goodwin tries to hack into AT&T’s server farm and is arrested by the FBI. That’s actually what a high performance computer looks like, below is a picture of the supercomputer ARCHER at the University of Edinburgh! But first, what is HPC?
ARCHER, EPCC’s supercomputer.
What is HPC?
First off, high performance computing is the aggregation of computers with the goal to achieve a higher performance than one would get from a single machine. If we view each computer (imagine your own laptop) as a node, a HPC system is a cluster of nodes each with its own operating system. So as I mentioned in the first blog, if we have a job to do, we can optimise the performance by sending different parts of the job to a different node. Alternatively, this is known as parallel computing.
How do you work on one of these computers?
It’s like working normally in the terminal. You log onto these computers remotely through ssh. For example, ARCHER (EPCC’s supercomputer) is 10km south of Edinburgh, at the Advanced Computer Facility, but you can work on it anywhere with an internet connection. So when working on the HPC system, you can write your programs in any coding language you want: C++, Fortran and Python are the popular options. However, optimising performance is always the end goal, so if possible, people tend to avoid Python because it is “cripplingly slow” from what I learnt in the training week. Furthermore, one must also deal with the added complexity of dealing with communication between the nodes that are being used, where we use MPI (Message Passing Interface) as the standard way to perform these routines.
Entrance to the Advanced Computing Facility. I was surprised it was on Google maps!
Where are these computers?
We actually got to visit Edinburgh’s Advanced Computing Facility and take a tour around. As you can see, it’s not a very welcoming site. We weren’t allowed to take photos inside because it’s seen as a security risk. However, they said they would supply me with stock photos of the supercomputers hahaha. It was a very thorough tour and we were accompanied by a group of people who were summer students here 20 years ago in the midst of their reunion. They were telling us stories about doing research on computers back then, they had no portable machines. So they have memories of doing all-nighters in the computer lab in order to get their work done. I don’t envy them!
Security is very high at the ACF, much higher than my home university where the smaller HPC system is on campus behind a locked door.
How much do they cost?
These computers are insanely expensive, ARCHER the latest supercomputer cost £43 million and its electricity bill is about £1 million a year. These computers are so powerful that the majority of the electricity is used for their cooling systems, otherwise fires would occur.
What work is done on them, and does it affect us in any way?
The final question to answer for this quick overview is what are examples of work done with HPC systems and will they affect our lives? Many branches of science and industry make use of them. Weather forecasting, Molecular dynamics, Aerodynamics, McLaren are doing research into developing a ‘hypercar’ using a HPC and the NHS use HPC to manage their database and analyse big data. In fact, we also had a seminar on how HPC is used for drug design. Martin Lepsik from CERMAV Grenoble described computer aided drug design and how they can perform virtual screens on drugs, simulate how they will bind with proteins and perform calculations to score the strength of these bonds. This saves lots of time which would otherwise be spent on clinical testings and enables them to try many more combinations than possible in a laboratory. I found it quite interesting as these scoring functions had a link to some of the work I did in my final year thesis involving Quantum correlation energies.
Picture of Jakub and Eva on our first day at EPCC, the James Clerk Maxwell Building is in the background.
Now the training week is over, we’ve began our projects and met our supervisors. Here is a picture of Jakub and Eva as we start our first day of work at the James Clerk Maxwell building.
Next post I will discuss my project. If you’ve any questions just ask in the comments and I will happily reply.
Doing a summer internship might feel like skipping a well-deserved vacation after a long year at uni, but fortunately most supervisors are pretty flexible about giving us students some days off. Last week I flew back to Manchester for a couple of days for my graduation ceremony, and even though I did feel slightly silly asking for time off only a week after getting to Jülich, everyone was very understanding. And it was definitely worth it! My family came over from Helsinki, and we were all in awe of the mysterious British robes and hoods everyone was wearing. Did ancient British students have three Simpsons-like fingers, or what is the story behind those sleeves?
After three years in Manchester!
After returning, I quickly got back to work. Three blog posts in, and I finally have something to say about my project! My supervisor and his colleagues here at JSC have developed C++ code that performs simulations of electron motion in graphene. You can read their paper here, since I don’t want to bore anyone by rambling about Hybrid Monte Carlo simulations and all that. The important idea is this: we have C++ code that generates descriptions of where electrons are in a graphene lattice at a given time, and how they are interacting. All of that information can be stored into a matrix (essentially a grid of numbers).
The fermion matrix for graphene is obviously a LOT bigger.
But the thing about computing, is that every branch of science has their own preferred programming language that they like to use to analyse data. It’s kind of a like a mother-tongue; speaking anything else seems to take a lot of effort for some people. Without trying to offend anyone (there are plenty of exceptions!), the stereotypical conversation goes something like:
Paula the Physicist: Python is great! It’s easy to read and write, so I can make nice plots of my data really quickly, and move it around in different files.
Matt the Mathematician: Oh no, Matlab is much better! You can write everything in vector-form, so that it’s very intuitive to read. I mean who doesn’t love to read things like V = 1/12*pi*(D.^2).*H?
Chloe the Computer scientist: What, Matlab? Python? Properly compiled languages are sooo much faster. C, C++ or Fortran are much better.
Fortunately, the eternal debate creates lots of small projects for us summer students. This week I’ve worked out how to save the matrix created by the C++ code we already have, so that it can be imported into Matlab. The idea is then to pass that data on to Matt, who will be happy that we are speaking his language, as it were. Now that we have this starting point, the critical question in High Performance Computing is whether we can do it faster. If you run a program on a small data set, a 10% speed increase does not matter much; it only makes you wait maybe a fraction of a second less. However, if you run the same program in parallel, using many computers and much more data, that 10% increase can become a 90% increase, and that fraction of a second can turn the code from something that would take weeks to execute to something that can be done in hours. More work for the rest of the summer then!
Hi, my name is James Lowe, I am 23 years old and I was born and raised on the north-side of Dublin, Ireland. I have just this summer completed my bachelors degree in Electronic and Communications engineering at the Dublin Institute of Technology. I am taking part in the Summer of HPC in Ostrava Czech Republic at IT4Innovations where I am working on a visualization tool for the performance of data.
Me myself and I
As part of my bachelor degree I was required to partake in a final year project. The project was to simulate the Ising Model of ferromagnetism. This looked at the change of magnetic state of a material with temperature and applied magnetic fields. The model required a HPC solution as simulation times would take far too long on consumer grade PCs. To combat this I built my own Beowulf cluster from old PCs and also applied for time on Ireland’s national supercomputer Fionn with the Irish Center for High End Computing (ICHEC). This was my first introduction to High end Computing and I really enjoyed it! From this, my thesis supervisor Dr. Kevin Berwick recommended that I apply to the SoHPC and here I am today!
The summer of HPC kicked off with a training week in Edinburgh, Scotland. Where David Henty gave a great introduction to Parallel computing using MPI. Edinburgh is a beautiful city and we got to see such sights as Arthur’s seat and the cafe where JK. Rowling wrote Harry Potter and also did a bus tour around the city.
The project I am working on in Ostrava is the visualization of performance data. There are many tools for measuring performance data such as Scalasca. They allow the programmer to optimize their code by showing the time a given function takes to execute along with other performance metrics. From this it is possible to make alterations to the code and improve its efficiency. However the feedback from the current tools are not intuitive. The aim of this tool is to take the data from these tools and produce a view on the data upon the communication’s model .
Ostrava is a beautiful industrial city with a lot of character. There are many interesting places here that i would like to visit. I had the pleasure of going to see the coal mines which i found interesting. The people at IT4inovations are friendly and welcoming and so far i am thoroughly enjoying my stay here.
After graduating my bachelors in Mechanical Engineering, the quest for finding the better possible version of myself began. I always had a passion for understanding the physics of objects around me. Fortunately, I found a job where I was able to participate in the building and simulation for a plethora of automotive assemblies, using conventional front-end tools, tested and validated under various criteria. Though fascinated in the beginning, I started to realize that it is more fun working on the back-end intricacies like solving a complex problem numerically and bringing out a visual appeal from the computed data.
My Master’s program in Computational Materials Science provided that opportunity by introducing me to multi-scale material modeling and computational mathematics through which I was able to gain insights on how to approach the art of problem-solving. In recent months, the field of machine learning was brought to my notice while browsing through the contents of material informatics. Also, it stipulated my level of curiousness to explore, how differently it works from typical simulations, where the equations governing pertinent phenomena are solved generally by conventional methods.
Fortunately, I found myself in a position to work as an intern in the PRACE Summer of HPC programme. I strongly feel that this opportunity can uplift my knowledge and confidence levels before starting my Master thesis and bridge the inter-space between my technical expertise and my shortcomings. My project is on the hybrid scaling of Convolutional Neural Networks, using High-Performance Computing.
St. Margaret’s Well, Edinburgh
The better start than I’d have probably asked for was the training week in Edinburgh. The lectures were on point, followed by a chance to run the code on ARCHER, the UK national Supercomputer. Along with classes on MPI and its applications, it was a fun-filled week. The spontaneous afternoon beach plan, the mid-week trek, restaurant hunts, mid-night walks, friends, hills, castles, a trip that will go down memory lane.
From the land of Bagpipes and convivial people, we took a cup of kindness and left to our destined sites. In my case, it is Amsterdam (Trust me! I’m not on a vacation). Until the end of August, I will be accessing Cartesius, the Dutch National Supercomputer at SURFsara, to run my simulations. In the coming posts, I will write more about my project. Until then, Goodbye from Amsterdam!
I am a 21 years old student at Poznań University of Technology in the field of Computer Science. In September I’m going to start my last semester of my BSc degree. In a nutshell, I’m a geography and floorball lover, never too tired of trying something new (isn’t life all about trying new things?).
Me and George during the bus tour around the Edinbourgh.
My programming journey started in the last year of high school. In the beginning I claimed that programming is boring, but then I saw some people writing code for calculation purposes. I must say that I got interested. Underestimating programming was a mistake, but also a valuable lesson. I learned how to discover things deeper. That was a turning point which changed my life plans.
During my studies, I try to be active in many fields. My main programming language is Python. I am also familiar with frameworks like Django for Python and jQuery for JavaScript. The big challenge for me is my Bachelor thesis. The aim of it is to classify some artifacts on crystallography photos using machine learning tools. It introduced me to concepts of neural networks, decision trees, regression algorithms and in general statistics.
I am a participant of machine-learning circle GHOST( check it out ! http://ghost.put.poznan.pl/en/ ). When our tutor Mateusz Lango showed me the HPC site I was very excited. I imagined how effective it would be to work with such huge computing power. I was inspired by supercomputers and that’s why I am a part of the PRACE Summer of HPC programme right now.
In the training week in Edinbourgh, day by day I gained more experience in C++ and MPI. Despite beeing familiar with MPI, I found out and learnt some really interesting information.
In the capital of Scotland was the place where I met people from many faculties. It was very inspirational to observe physicists, mechanical engineers and computer scientists developing their code. Moreover, I had a chance to experience many foreign habits like the ‘irresistible need of keeping sunglasses with you wherever you are’ (thank you George). I couldn’t believe that someday I will use to word ‘lunch’ ( in Poland in general we don’t use it), but now it’s in my blood. My family tried to help me gain weight from many years, but what they really should do is send me to Edinburgh to have some lunches. It helped. Really.
It took me a bit to get used to left-hand traffic in UK.
All good things must come to the end. From Edinbourgh I flew to Ljubljana where I started a new adventure with HPC. You won’t belive how beautifull this city is…
At the Univeristy of Ljubljana I am working on data about radiation levels in the world using Hadoop and R programming language. Hadoop is a framework that allows for the distributed processing of large data sets across clusters of computers using a simple programming models. Hadoop works with assistance of the MapReduce algorithm. It allows me to get the most important information from big data (i.e txt files of size 20 Gb).
I am really glad that I have a chance to meet the world of HPC!!! Who knows, maybe one day you will be able to say the same words !
(Apologies as this is quite late in being posted, as I’ve had a few hectic weeks, with moving accommodation, getting started on the project and just adjusting to living here.)
Enough excuses, back to the blog post.
Well, my name is Conor O’Mara and I am 22 years old (this is giving me secondary school French oral flashbacks ahhhh!). I am from Dublin, Ireland and have just finished a Bachelors in Theoretical Physics at Trinity College Dublin. I have been lucky enough to receive the privilege of being awarded a place on the PRACE Summer of HPC programme where I will be based at the University of Edinburgh’s Parallel Computing Centre (EPCC).
During my final year thesis I was given some exposure to High Performance Computing (HPC), where I used Trinity’s Kelvin cluster (named after Lord Kelvin, the Scot-Irish scientist who has the absolute temperature scale named in his honour – The Kelvin scale) to perform some calculations. These calculations were in the field of Quantum Many Body theory which I undertook with the supervision of Professor Charles Patterson. While I was doing my final year project, another final year student called Dovydas Mickus, who also had the same supervisor as myself was working on a project that focused on parallelising some of the code for calculations in order to increase the efficiency and speed of these calculations in the allocated node time. As we met as a group to discuss our projects with our supervisor, this was my first exposure to the idea of parallelising calculations across many nodes (ie. sharing the job by splitting it up into many pieces) to speed up the calculation. So I have both of those guys to thank for my introduction to HPC and parallel computing.
Here is a picture of (L-R) myself, Dovydas Mickus and Prof. Charles Patterson.
Picture (L-R): Me, Doyvdas Mickus, Prof. Charles Patterson.
In Edinburgh, I will be working on a project which simulates and tests different job scheduling algorithms for how a HPC allocates nodes. The project will encompass some software engineering, testing and some work with algorithms. It is quite a computer science heavy project, which is a little daunting for me and my physics background. However, I am looking forward to the challenge as I view it as a good opportunity to develop my skills and further my knowledge in this field.
Other aims of this blog post are to communicate my experience during my time here and also advertise this program so that hopefully more people from Ireland will apply to the PRACE Summer of HPC programme. There are 23 people from across Europe who have been accepted to work across difference locations in Europe. 3 of us are based here in Edinburgh, while different people are in places ranging from Nicosia, Bologna and Barcelona. PRACE fully supports us with a stipend for accommodation, academic supervisors to help us with project and living expenses. So far, I would highly recommend you apply if you are from a Physics, Chemistry, Maths or Computer Science background.
The program began on the first week of July where everyone on the program met in Edinburgh for a training week. The week itself was great, the standard of teaching in the university was very good and I got an nice overview of HPC that knitted in well with my current knowledge as well as covering lots of new material I had never seen before (perhaps I will discuss about a subject or two in a future blog post). But the emphasis was shared between social elements too and we ate out together most nights, visited most of the main tourist attractions and most importantly had a laugh the whole time which is the type of laid back environment I like to work in.
Here are some photos from the first week featuring an aesthetic African restaurant (thanks to Chef Mo for having us), a Rick & Morty themed bar, Edinburgh castle, the group on top of Arthur’s seat and some other gems.
I’ll leave you until next time with a quote from the famous Edinburgh writer Sir Arthur Conan Doyle, which will stand as a reminder for me to blog more.
“A trusty comrade is always of use; and a chronicler still more so.”
It’s going to be an awesome journey fellas, and yes the title could be made into a movie!
Hi, my name is Atul and I am 25 years old. I hail from India, and I am currently enrolled as a MSc student in Computational Science and Engineering at the University of Rostock, Germany. These are (and will be) a set of posts on my thoughts, experiences and work, as a participant, of the PRACE Summer of HPC 2018 programme, where I’ll be working on a HPC (High Performance Computing) project for 2 months at one of the supercomputing centers in Europe. So come with me as I embark on this journey and take you through one of the most amazing experience of my life.
Not so long ago, I was building cars and all terrain vehicles as college projects couple of years back.
The Invader, a project I was a part of during my Bachelors. Looks Familiar, right?
That was done during my Bachelor degree in Automobile Engineering. I’ve now developed, as a participant of this programme, to build on a small infrastructure for HPC applications (think software). This is again inline with my Masters’ as it explores the simulation aspect of engineering and sciences.
How did I got here? I suppose, it has more to do with the positive cosmic forces that drive our universe. I may have caught and synched with an appropriate frequency.
The PRACE Summer of HPC programme is one such beautiful instance of an “event” (Get it?) which came my way and each day has been absolutely phenomenal so far. I’ve had the privilege and blessing to visit the beautiful city of Edinburgh along with 19 other participants coming from various European universities. Oh and did I mention, I am also going to spend the better part of this summer in Bologna, Italy. Not everyday does a student get such an amazing opportunity to work on a challenging project, and be able to see Europe as a detour from work. Surely the makers and organizers of this programme deserve applauds for thinking this through, or at least a big shoutout from one very happy participant (or perhaps 20, each year)!
I remember the conversation that I had with myself when I applied for this. The guy with the thorns on his head saying “Why bother! You won’t stand a chance against all the PhD’s going for this, that are from across European institutions, which is as good as saying , students from all over the world.” His demeanor was understandable as for more than half of the projects open for applications, I understood only very few of them and was eligible for only a couple of them. However, the guy with the ring on his head said, “Look, dude, if you won’t apply you will definitely not get it, but if you “just” tried, you may probably get it. That is statistically more likely, if not highly unlikely”(Yeah, my subconscious talks in probabilities sometimes)! The choice was obvious. I had to give it a shot!
I was busy running some errands for my brother’s wedding (yeah, that was in April of 2018), when I got accepted. The day just became brighter, more lively (as is the case in most of Indian weddings) and of course even more fulfilling because I was already in a place, both emotionally and physically where I could actually celebrate with my whole family. My sister seemed more happier than me, “Wow, you’ll be seeing Italy, now eh!” she says. To which I reply, “I’ll be living there, sista!”
I couldn’t have asked for more, every aspect of it has been amazing so far. The design is just spot on, accompanied with the apt resources and infrastructure to make it such a brilliant success between both participating individuals/company. The training week in Edinburgh gave me (and us) a taste of power of a supercomputer and what it could do when used properly (and also, how my laptop is not that powerful, as I once thought as a kid while buying it).
The awesome training week, at the awesome city of Edinburgh, with an awesome supercomputer, awesomely named Archer with other fellow awesome participants!
The exercises during that week introduced and educated us on how we as a participants should be leveraging these powerful resources to the best of our abilities for the projects that will be assigned to us.
Post classes, there was of course, Edinburgh to explore. Just look at this animation google made for me, it will basically give you a “feel” for it. Spoiler alert, that’s Edinburgh in the backdrop when the seagull takes the epic nosedive!
I then, after the completion of the training week, flew to Bologna, where Cineca, a supercomputing center and my placement site, is located. I must say, there is something new to learn, every single day. As much as it sounds cliché, getting to do the real thing, beats all classroom lectures. Even my classroom projects look cute when I compare them with what I learnt in the 1st week on one partition of a fully fledged supercomputer. There is simply no comparison.
I’ve been invited to work on a CFD project that is currently, and has been for quite a while now, a hot topic of research among the research community. (Think CFD as simulating and learning about the behavior of fluids interacting on or with some object, which can be air, water or any other fluid. There are of course other complexities involved). This has importance especially because large part of it depends on computation that should produce results as quickly as possible. Hence making the supercomputer extremely useful in such an endeavour. The applications are many, for e.g. The airplane takeoff leaves a vortex so strong behind it,that the other plane behind it must takeoff after this vortex diminishes. Understanding this vortex for example, could help to better organize the take off times of flights at airports, perhaps increasing the airport output in terms of revenue.
Simply put, my project would be to create programs that help visualize similar such computations, while it is being computed on the supercomputer! So think, saving resources on a machine that saves time. There is a also a word for it, “In-Situ Visualization”. Sounds awesome, right! Well, that’s because, it is.
Ladies and Gentleman, presenting , with 20 PFlops/s, featuring more then 300,000 cores, the one and only. Marconiiiiiiiii
Not to mention that I get to do it in Cineca, which is currently the world’s 18th most powerful supercomputer (Check here the top 20) in the world.
I simply feel blessed to be ableto work on it with such amazing individuals helping me in my learning curve. These people running such centers are obviously the best in their fields (which is most of the time, Mathematics with honors or magna cum laude, I’ve observed). This simply means that the water cooler moments also become interesting, engaging and of course fun (nerd jokes are not only gregariously appreciated, but also almost always responded with another one).
Well I know by now, you must be totally intrigued about this, but cliffhangers are always awesome as they, well, leave us hanging on a cliff!
So that’s all for my introduction for now and for this post, stay tuned as I’ll keep publishing the goodies in the time ahead as this project develops and this beautiful summer blossoms. For this is, and is going to be a “Summer of Highly Piquant Circumstance” of mine!
Hi everyone! My name is Pedro Santos, and I’m participating in the PRACE Summer of HPC 2018 programme.
I’m from Coimbra, Portugal, home to one of the oldest universities in Europe (dating back to 1290), which has very rich and unique academic traditions, which I carry in my heart. I have obtained a degree in Chemical Engineering at the University of Coimbra, after finishing my MSc thesis in September 2015 in the area of polymer science. Afterwards, I worked for a year in that field, but I grew dissatisfied with the routines of a chemistry laboratory. I wanted something more challenging.
That’s when I was introduced to the exciting world of molecular modelling. It became a way to express my love for chemistry and computer science at the same time. This is a rapidly growing field with immense opportunities for research. Since March 2017 I’ve been working on molecular simulation for materials design, as part of a research project, and I’m starting my PhD in this area next year, at the University of Coimbra.
It was in this period that my supervisor first told me about PRACE Summer of HPC. As I was still taking my baby steps into molecular simulation techniques, I immediately thought this was the perfect opportunity to take a big step forwards in my learning curve. On one hand, I get to learn about High Performance Computing (HPC), the “beating heart” that drives all chemical simulations today (and without it the exponential growth in that field would have been impossible) and also, as a bonus, brush up my programming skills. On the other hand, being accepted in a project that involves the use of molecular dynamics and metadynamics, two techniques I will extensively use in my PhD, is an amazing opportunity to get expertise from people with much more experience in those particular areas.
PRACE Summer of HPC also presented somewhat of a personal challenge to me. This was the first time I traveled abroad and by airplane. At first I was afraid, but I took on the spirit of the Portuguese discoverers of 500 years ago and surged on into the unknown. I actually had a lot of fun in all my travels, and even got a sneaky peek into Paris in my first connection. Still, I would say I prefer the quiet and cosy setting of Edinburgh, which I had the pleasure to explore in great company during the training week. It was an intense but rewarding week, as I flexed my programming muscles with MPI, and shared stories with wonderful people from around the world.
And now, as I settle in Athens and in the lab supervised by Dr. Zoe Cournia at BRFAA (to whose members I thank for the warm welcome), a new adventure is beginning. I will be working on an exciting and extremely relevant project, regarding the simulation of a mutation in the PI3Ka protein, which is known to be related to carcinogenesis. I’m sure I will learn many concepts and expertise that will be useful for me in the future. I will also have to face the daily challenges of living in a foreign country, with a language unknown to me (even a completely different keyboard layout! o.O) – but that will only make me a better man in the end. And, most of all, I will explore the best this city has to offer and have some fun!
A photo I took of the most iconic landmark of Athens: the Acropolis!
For now, I leave you with a fitting piece of wisdom, that emerged from this country 2500 years ago:
“There is nothing permanent except change.” (Heraclitus)
I am a student of the School of Electrical Engineering, University of Belgrade majoring in Software Engineering. The Summer of HPC programme was introduced to me by my Professor who attended it a few years ago, and who thought it was an awesome experience. After looking at the offered projects, I got pretty excited, as most of them deal with real problems and data, and on a large scale too.
So far, the people I’m with in Ostrava, which is where I’m situated, have been nothing but amazing. I think I got lucky with my mentor, Martin, as he’s very responsive to any questions I have, whether it’s about the project or simply getting to know the surroundings. He’s away for a few days now, but I’m getting him a beer as soon he comes back. The other student, James, who’s with me in Ostrava for the summer is this really laidback guy – extremely easy to talk to. So yeah, I’d say I’m in pretty good company for the weeks to come.
James and I in our IT4Innovations working environment.
As for the project, I’m dealing with software that assembles genome from a large amount of short DNA sequences. In other words, there’s lots and lots of data and it can take days to finish processing. This is where I come in – my main job is to improve the efficiency of the implementation, whether by making small improvements with DNA file readers or making sure that as many stages of the assembly are running in parallel, multithreaded and multiprocessed. The whole bioinformatics field is absolutely new to me, so initially it took me a bit of time to get myself acquainted with the mechanics of algorithms being used and various new terms. I’d say it has paid off, though, as the approaches used to handle DNA data are rather clever, particularly from an engineering perspective.
As far as I’m concerned, this is a good start and I’ll be updating with new information as the situation progresses.
I am Eva, a Numerical and Computational mathematics student from Prague, Czech Republic.
My main field of interest is numerical linear algebra, specifically all kinds of matrix calculations.
Just recently, I started to focus on discrete inverse problems as a part of my Masters thesis that studies algebraic methods for single particle reconstruction, i.e. particle shape reconstruction from projections. Outside of the university, I enjoy playing sports, mainly Ultimate Frisbee (which is not just a random throwing of the frisbee in a park!), puzzle solving, travelling, eating chocolates, cooking and flipping pancakes. I am slightly addicted to coffee and my favourite movie character is Dave the Minion.
This summer I will be working at the Edinburgh Parallel Computing Center (EPCC) on Parallel Computation Demonstrations on Wee Archie. To start with, let me share my impressions from the first two weeks here.
Besides getting to know the other participants, the first week was dedicated to training and mainly focused on an introduction to HPC and programming with MPI. The highlight of the week for me, however, was a spontaneous afternoon trip to the Portobello beach (You just cannot leave a seaside city without going to the beach, right?!). It always sounds promising when a group of 9 young people from all around the world who barely know each others names decide to do something adventurous. The take-home message from the trip is that if a guy from India leaves the group claiming he will be back in 2 minutes it means that, if you are lucky, he will turn up sometime during the next half an hour, maybe.
Portobello beach. Thanks Wojtek Laskowski for the picture.
The plan for the second week was to slowly get to know our mentors, colleagues and to settle down at the EPCC and in our new accommodation at Haymarket. Unfortunately, the actual plan for me changed to getting some medicine and resting in bed. It seems to take some skill to catch a flu during a heatwave in Edinburgh but yeah, it is possible. By the way, did you know that a pharmacist in the UK might not actually speak English at all? Pantomime, however, seems to work anywhere in the world.
My name is Petteri. I am from Finland, where I am currently finishing my MSc in chemistry.
I started my studies in the university of Turku in 2013, studying geology. After taking my introductory course in physical chemistry, I decided to pursue for a degree in chemistry (although I did finish my BSc in geology on the side because rocks are nice).
My initial plan of specializing in physical chemistry changed in 2015, when I joined David Palmer’s research group in Strathclyde University for a few months. After three exhilarating months of molecular dynamics (MD) and supercomputers, I knew there was no turning back. I returned in 2016, this time to learn the (very) basics of 3D-RISM and to deepen my knowledge of MD.
Late last January I received an email from PRACE, which advertised the Summer of HPC program. Although 91% of all the project topics looked alien to me, I could spot two topics related to computational chemistry. I decided to apply for project #14, named High throughput Virtual Screening to discover novel drug candidates. The motive here was quite simple: I had done a bit of classical stuff and a bit of non-classical stuff. Virtual screening was (and still is) an area where I have no experience whatsoever.
Late last March I got the good news from the program coordinator – I was selected. Fast-forward a few months, and I found myself in Edinburgh participating in the programs starting week, which was enjoyable. Made a few friends, learned quite a few new things about message-passing programming, and enjoyed a few nice local lagers in good company.
After the training week I flew to Athens, where I started working in Zoe Cournia’s group, based in the biomedical research foundation of the academy of Athens (BRFAA). During my stay here I will be mostly focusing on using virtual screening methods to try and find a isoform selective inhibitor for an enzyme known as ALDH7A1 (aldehyde dehydrogenase, seventh subfamily, member A1).
ALDH7A1 and its first solvation shell.
The motive for this project stems from the fact that overexpression of ALDH7A1 (with other isoforms such as ALDH1A1, ALDH1A3, ALDH2, and ALDH4A1) has been found in multiple types of cancer. In these cancers ALDHs promote drug resistance and amplify the amount of cancer stem cells (a group of cells which can initiate and propagate tumors, among other things).
To remain impartial, I should probably add that the ALDH family members also do non-cancerous things which include, but are not limited to, catalyzing reactions which detoxify reactive aldehydes in the body (such as acetaldehydes), and help reduce oxidative stress.
Stay tuned for more updates and cool figures!
Hen egg-white lysozyme – just something from one of the tutorials I was working on earlier. The figure shows how the protein is moving throughout the simulation.
Further reading
C. Chan, J. W.Y. Wong, C. Wong, M. K.L. Chan, and W. Fong: Human antiquitin: Structural and functional studies. Chem.-Biol. Interact. 2011, vol. 191 (1–3), p. 165-170.
C.van den Hoogen, G.van der Horst, H.Cheung, J. T.Buijs, J. M.Lippitt, N.Guzmán-Ramírez, F. C.Hamdy, C. L.Eaton, G. N.Thalmann, M. G.Cecchini, R. C.M.Pelger and G.van der Pluijm: High Aldehyde Dehydrogenase Activity Identifies Tumor-Initiating and Metastasis-Initiating Cells in Human Prostate Cancer. Cancer Res. 2010, vol. 70(12), p. 5163-5173.
M. Luo, K. S. Gates, M. T. Henzl, and J. J. Tanner: Diethylaminobenzaldehyde is a covalent, irreversible inactivator of ALDH7A1. ACS Chem. Biol. 2015, vol. 10(3), p. 693-697.
V. Vasiliou, D. C. Thompson, C. Smith, M. Fujita, and Y. Chen: Aldehyde dehydrogenases: From eye crystallins to metabolic disease and cancer stem cells. Chem.-Biol. Interact. 2013, vol. 202 (1–3), p. 2-10.
When I was a kid, we only had my brother’s computer at home. I was 5 years old and I was only allowed to play with the computer for 15-20 minutes every day. Then one day, my brother asked me “What do you want to be in future?” and I answered “Computer Engineer! So, I can play with the computer whenever I want.” and this is how a dream started. I wanted to be a Computer Engineer since childhood.
On the Hallstatt Skywalk
My name is Enes, I’m from the beautiful country of the Republic of Turkey. I just graduated from my BSc in Computer Engineering. I’m planning to proceed with a Master’s degree and PhD outside of Turkey. But, apparently I’m not the only with the same plan, so finding a scholarship for a Master’s degree is challenging. Consequently, I’m planning to skip a Master’s degree and begin with a PhD directly because it is easier to find funding.
I see myself as a lifelong student. Because, I have a hunger to learn new programming tricks and I really enjoy programming as a hobby. I quickly get bored when I work on the same thing for a long time, so I have looked into different areas for years. Also, I love competitive programming contests, hackathons and have won prizes in these. Until the previous year, I was working in web technologies as a full stack developer. I’m experienced in many web frameworks and different languages such as: Spring, Django, Node.JS, ASP.NET and so on. I then started a Machine Learning course by Andrew-Ng and that course changed my future plans. Currently, I’m focused on Machine Learning and Neural Networks. I also play with Blockchain kind technologies in my free time.
Last semester, I took a Parallel Computing course at my university by our tutor Dr. Mete Akdoğan. We learned about memory architectures, libraries such as OpenMP and MPI, Big Data concepts and so on. We also used a cluster for the first time, the cluster of the National High Performance Computing Center of Turkey (UHeM) which is the PRACE partner in Turkey. Our tutor also told us about the PRACE and Summer of HPC programme. I then started to check the available projects and saw some which were related to machine learning. I was really excited when I applied and while waiting for the decision. SoHPC is a really good opportunity to meet new friends from new cultures and meet future scientists, to visit at least 2 countries (for most of us) and work on great projects which are designed to teach us many things.
I was surprised by the program of the training week in Edinburgh – they were really prepared and everything was on right track. There were many exercises for everybody from different programming levels. Even though I followed a similar course at my university, I learned new things in Edinburgh. But, my favorite part was walking to Arthur’s Seat.
On the Arthur’s Seat with Nazmiye and Zheqi
Last but not least, I’m very glad to be in this program. Currently, I’m working on implementing an easy machine learning K-Means algorithm with Global Address Space Programming Interface (GASPI). Last week I implemented it, then parallelised it with MPI and currently I’m trying to figure it out GASPI implementation GPI-2. So far, I enjoyed my project and learning new technologies.
I recommend Summer of HPC to everyone who wants to learn more about HPC and meet awesome people !
My name is Wojtek, I’m 25 years old and thrilled to announce that I’m spending this summer in Barcelona, Spain. Besides leaving home for 2 months for one of the most fascinating places in the world, I’ll also be working on the Adaptive multi-partitioning for the parallel solution of PDEs under the supervision of researchers from Barcelona Supercomputing Center.
But first things first, let’s put all the excitement on the back burner (for a while). Here’s a short description of myself and how I ended up in the heart of Catalonia.
Photo taken during the training week in Edinburgh. Trying to make the best use of my free time and find the perfect view of the city.
I am originally from Lublin, a small city in the South-East part of Poland. Quickly after finishing high school I moved to Warsaw, the capital city, to pursue a Bachelor’s degree in Mechanical Engineering. For my graduate degree, I sought a bigger challenge. I decided to move to Copenhagen, Denmark and start Master studies in Applied Mechanics at Technical University of Denmark.
Ever since my undergraduate studies I was interested in simulating engineering problems (fluid flow, plasticity etc) and numerical analysis. But it was only during my Master studies that I started to explore more and more what’s within the engine of computer simulation software and slowly exchange engineering classes (bye, bye commercial programs) for more specific, mathematical and programming orientated courses (hello Unix shell).
This is where I stand right now, finishing my master thesis, in which I try to design an efficient iterative strategy to speed up nonlinear wave simulations. With (hopefully) successful completion of the thesis I hope to make my own small contribution in the analysis of ocean wave propagation and wave-body interactions.
My interest in Computational Fluid Dynamics was the primary reason I signed up for the PRACE Summer of HPC programme. I believe that my new experiences in parallel programming can make my contribution a little bit bigger, as it allows for even faster simulations. On my way here I only discovered more reasons to join the event. Visiting Scotland for a week with a bunch of very smart individuals and then spending the summer in the most popular Spanish city while working on an interesting project doesn’t sound bad at all, does it?
Another photo from the Training Week. Finally free after whole day of programming.
When I am not pondering on a math problem, I either play or watch football with some breaks to look after my early cactus plantation (consisting of 2 cacti so far) and play board games with my friends in an unusually competitive manner. However, all of this will now be set aside for 2 months in favor of exploring the Iberian Peninsula in greater depth.
In the gardens of the Jaume I University in Castello de la Plana, Spain
Hello fellow reader,
Let me introduce myself – which is the most boring part in this post, but I promise after that it gets interesting. My name is Klajdi Bodurri and I am 23 years old. I was born in Albania and I grew up in Greece. Right now, I am an undergraduate student at the University of Thessaly in Greece. I am studying Electrical and Computer Engineering for 4 years and next year I hope I will get my diploma (fingers crossed). My field of interest is Computer Science and specifically developing algorithms for Distributed Systems. Currently, I am in Castello de la Plana, Spain for the PRACE Summer of HPC programme. I am enjoying the peaceful life here and working on a very interesting project (I am gonna tell you more about it later, keep reading).
But why did I apply for Summer of HPC? I think the answer to this is the curiosity. Curiosity, this little tiny thing that has built the whole scientific community. For me, it started in the summer of 2015 when I participated for the first time in Research and Development at my university and my first impression was “Cool, I should do it every summer”. Next summer, in 2016, my friends and I started building a prototype of smart auto locking system for bikes, just for fun. In the summer of 2017, I did my first internship and I saw how a start-up company works and how to be a part of a huge team. So, this summer, guess what? Summer of HPC!
To be honest though, fulfilling my curiosity wasn’t the only factor for applying in Summer of HPC. I was really lucky because last semester I was an exchange student in Finland. I met people from all around the world, I learned how to live abroad, how to interact with people from totally different cultures from my own and lastly and most important I learned how to survive in -30 °C. So, while I was in Finland, I was trying to find ways to expand the period of living abroad. In that moment, a friend of mine texted me about the PRACE Summer of HPC programme and I read that this is a programme, that offers summer placements at HPC centers across Europe. I said to myself “Time to learn more about High Performance Computing and meet new people. Let’s do it!”.
And here I am now in Castello de la Plana, Spain working on the project ‘Dynamic management of resources simulator’ which will simulate the execution of a workload composed of malleable and non-malleable jobs, over a parallel system. Other than that, I am enjoying the Spanish life. Time to take my siesta!
