 Hello. I think you should be able to hear us now. Yeah, hello. Welcome to everyone that's here. Yeah, today we're starting the second day of Thursday. How do you want to calculate it? The second day of the course, where we now start to focus on this HPC side. Let's talk a bit about the first day. I think we can talk about that. So, yeah, at the end of the stream, we forgot, well, we left it a bit at the end, that we wanted feedback from you. So well, maybe, maybe at the end of the day, when you we ask the feedback for this day, maybe you can also comment on the first day if you have additional feedback. But you can already on the first day. During the icebreaker here, you can already add some comments from yesterday if you had them. So yeah, because like, like everything in life criticism makes you better, like if you don't know what to what is the thing that you're striving for, you do not get better. So we appreciate feedback on everything good and bad, like what was good, what didn't work. So the HackMD is the same as yesterday. If you can open that up, we have these initial questions there. Icebreaker and so on. Yeah, yeah, today's icebreaker is related to, well, what we are going to be running there. Like, like we talked yesterday that the SPC system is kind of like a remote location where you go, like, like, it's a different place where you do different kinds of things. So similar, it's like, you, well, nowadays, your office might be your home, as with us, me and Richard. But at the same time, if you go to office, you usually do different things that you do at home. And similarly, with HPC, you might do different things there that you do at your normal office. So it's like a specialized place for specialized things. So when you go there, you usually want to pick something, some application or something that you want to do there, or some specific case that you need solve that you need the SPC resources for. And in the icebreaker, we have a question on what do you need to run your work? See, what did you think of the overall flow? What did you think of the overall flow of yesterday? Yeah, I liked it. Yeah, I it's always complicated when you have this kind of like the idea behind yesterday was that we would have this kind of like introduction, gentle introduction to the concepts, because not everybody's familiar with the concepts. And then we would give some heads up things because like this limited amount of time. And I think that, like, I liked what we had, but at the same time, it's unfortunate that we didn't have time to go into actual like examples of many other things that we were talking about. But at the same time, like I for some advanced users, it might have been too generic and too philosophical and too much old stuff. But it wasn't it was designed to be for the users who maybe aren't familiar with the concepts. So there and there. It's hard to you never you never can fit everything into one day. How about you? Yeah, like, previous years, there had been the complaint that we'd like go straight to things and then we get to the end and people are like, is this even the right place for me to work on stuff? Oh, but I can't make this program run. Or it's like, oh, but this program doesn't even run faster on the cluster than my computer. And why well, because it's not really designed for it. And there's certain kinds of applications that really scale well, or people who like start working there and it's too difficult or too weird, too different. So being able to like frame what you need before you start on this path was really helpful. I mean, I found it interesting that one feedback from yesterday was that it's good because it was very comprehensive. And one feedback was, you should go straight to the point more quickly. What do I think? Yeah, yeah, that happens to most of our courses. So I mean, we can't make one thing that suits everyone. But we try to make it clear what days are useful for what people and make it easy for people to skip over the stuff that's not needed for them. Yeah, like, like this is kind of like a balancing act between like, going too much into the specifics when when you might get a feeling that, okay, well, like this is not relevant for me because like, I don't use this tool or I don't, this is not relevant for me because I'm from a different university and you're talking about all the stuff on only and like, this is too specific to like small, like to too much detail. And then there's like the other end, which is like, this is so big that I don't even understand what's happening anymore. So yeah, there's a phone ringing in the background. I'll have to go for a second. Yeah, like yesterday, we were really enthusiastic about like, what you were asking in the HackMD, it was really great. It was excellent conversation started because like, for and also like, give us a better glimpse of what what's your reality, like, especially now in the COVID days, when we don't have that much physical interaction, we can't see ourselves with the users in the offices or in the coffee rooms, which was basically the place where lots of the best discussion happened. These kinds of like HackMD's type situations where you can give you all well, anonymous, direct, like, contribution. It's, it's it's very refreshing. And it helps us a lot. And it helps us focus on when we are talking to the right subjects. So it's very much appreciated that you can like, you have asked the questions. And especially today, when we are going to go a bit more into the detail, there might be some concepts that we, like, we explain in a certain way, because we have used it, or we have looked at it in that way for a long time. But that point of view might not, it might not, well, be easily transferable to you, it might be that we word it incorrectly, or we word it ambiguously. So it's, it's hard to decipher. So so asking the HackMD questions, let us give us a way of re-evaluating our words and, and change our way of teaching so that you can better grasp the concepts. Yeah. So here, I've added a few polls to the HackMD. So if you come down and add another letter to it. So this isn't a add a, like change the number, but go like this. So add in another another O to make a bar graph. So especially this first question, I'm connected to a cluster. So we're assuming that since we ended yesterday, you've been able to get connected. And I've been able to debug it and ask for help. So we're not really going over things again. Yesterday, also, we had various questions from various people all around the world, actually, asking, like, access, like, can they get access to, to our cluster? And unfortunately, that's, that's not possible. We don't have unlimited resources. So it's, it's for the university of staff and researchers only and, and people who do research there. But, but I would highly encourage, like, seeking out the local cluster that you have or collaborators like this. If you have collaborators who have access, they can probably get you access to wider array of clusters in EU or everywhere around the world. But even if you do not get the access, like, if you stick around to the course, we can still teach you the things because like, the same concepts are basically universal, all of the clusters in the, in the world use some kind of, like, or like, at least, or may most of the HPC clusters use some kind of a key system, similar to what we have. And, and the same concepts apply. And, and even if you feel like, okay, it's not fun to do the course now, because I cannot do the exercises, you can always, always look at the videos that we post on on our YouTube channel afterwards. Richard, I think, already made the previous day available available yesterday. Yeah, so it's, yeah, so it's already posted. We try to do that the same day, but also Twitch immediately archives it for 14 days. So you'll have to search through the right sections in there. But um, yeah, I don't see enough yeses here on I'm connected to a cluster. So I see around 60 people on the stream right now. So please start up and answer the poll. Yeah, because that will become like, like, if you want to run on any of the examples, if you want to test out the things yourself. And now we're starting the hands on part really, like we try to make it so that you can run as many of the examples that we are going to be presenting. So it's good idea to like, get familiar with the environment in the cluster by running the commands yourself. Because then you always will get instant feedback if there's something wrong, like if you if you type something wrong, and you get accustomed to the command line interface by actually typing the things. So if you haven't yet connected, now it's a good time to connect. There's like, I mentioned yesterday, there's in the HackMD at the top of it, of the document, there's like this recommended, all recommended layout, if you want to put your window and terminal and everything like that in different places so that you can even on one screen, one full HD screen, you can probably have a terminal open and a stream open at the same time, hopefully. Yeah. Should we summarize the day I guess everyone who's coming is likely here. I have our schedule open, which I will share. Yes, here it is. So let's see. Today, what are we discussing? So first we discuss about what you can do with the cluster. And then we quickly go to two real examples. So this is sort of a inversion of what we've done previous years. Other years, we start off very basic and then we show all the pieces and then we show everything put together. So now we're going to start by showing you everything put together as a demonstration. So basically, we'll go too fast for most people to be able to follow. So you'll just sit back and watch and see. And then we'll go over all the little steps that go into that so that by the end of tomorrow, you're able to do all of these things. Um, yeah, and then after that basics, first we'll talk about actually running the jobs. So the slurm batch queuing system that connects things together. Then we'll talk a little bit more about environmental things like how you find software on the cluster and how you store data, which are actually important. So before we would talk about things like accessing the software before we talked about running jobs, but that made it a little bit abstract. So it's really like an immersion today. And then tomorrow we go to the like parallelization kind of thing. So basically being able to run multiple things at once in three different methods. In addition to a talk from you see from CSC who talks about some of the even larger resources available at a national level. So with that being said, let's take a look at the HackMD again. Okay, we're getting more answers here. More people are being connected. Yeah. Um, should we begin then? Um, yeah, let's see. So first time the schedule is about clusters and our work. So maybe we can head there.