Blog

Python in High Performance Computing 27.1. - 29.1.2020 ***

In this course we teach various ways to optimise and parallelise Python programs. Among the topics are performance analysis, efficient use of NumPy arrays, extending Python with more efficient languages (Cython), and parallel computing with message passing (mpi4py) approach. This free of charge course is part of the PRACE Training center activity at CSC. Read more and register (by 17.1.2020) at https://www.csc.fi/web/training/-/python-hpc-2019

Webinar: Data migration from Taito to Allas (1/2)


Time: 13.11.2019, 13:00 EET

Lecturer: Kimmo Mattila, CSC

Language: English


https://www.csc.fi/web/training/-/webinar-data-migration-taito-allas


Taito.csc.fi will be closed by the end of 2019. All data that needs to preserved from the disk environment of Taito, needs to be moved or copied away from Taito before that.


This webinar shows how you can back up your important data sets from Taito to the new Allas object storage service and how you can use this data in the new puhti.csc.fi cluster.


Join the webinar in Zoom: https://cscfi.zoom.us/j/102837163

FCAI arranges a series of events about AI and X, where X is something else, a scientific discipline, an application area, another Academy flagship etc.

 

Please follow the FCAI webpage for information: https://fcai.fi/aix-forum.

 

The next events are as follows:

 

Friday 1.11. at 11-15 in Turku, AI + Health and Maritime, arranged with Turku Science Park, for registration and further information see https://turkubusinessregion.com/tapahtuma/ai-maritime-and-health-with-fcai-visit/

 

Monday 4.11. at 15-19 in Dipoli, AI + Economics and Causal Reasoning, arranged with Helsinki Graduate School of Economics, for registration and further information scroll down to the event on https://fcai.fi/events.

 

Wednesday 27.11. at 13-16 in Porthania, an event in Finnish about AI + Public service (Tekoäly ja julkishallinto), arranged with Palkeet, for registration and further information, see the attachment (in Finnish). [Note: somewhere there is a wrong date going around, the right date is 27.11.]

*** Scientific visualization with Blender 5.12.2019 ***

Blender is free and open source 3D creation suite, that supports the entire 3D pipeline. This course is an introduction to Blender 2.8, focusing on features useful for scientists. Hands-on exercises and demos cover the basics of Blender. Registration deadline: 22.11.2018. Read more and register:
https://www.csc.fi/web/training/-/blender_2019

Introduction to Parallel Programming (23.10. - 25.10.) *

The course introduces the basics of parallel programming with the
message-passing interface (MPI) and OpenMP paradigms. The course
consists of lectures and hands-on exercises. This free of a charge
course is part of the PRACE Training activity. Read more and register
(by 16.10.) at:
https://www.csc.fi/web/training/-/intro_parallel_program_2019

Time: 21 October, 9–11am
Place: Aalto University, Computer Science building, T1

Have you ever wondered how to harness the power of artificial intelligence to benefit your own research?
Do you wish to collaborate and create something new and exciting with top AI researchers?
Do you want to know more about the flagship program of the Academy of Finland?

FCAI organizes a community event on 21 October 2019 to share what we do and to welcome new researchers from Aalto University, University of Helsinki, and VTT to join the FCAI community. We invite AI researchers as well as researchers from different fields to join the event!

Please register at https://link.webropolsurveys.com/S/24031E794BBD7450


Program (subject to minor changes)


9:00: Coffee
9:15–9:45: Finnish Center for Artificial Intelligence: excellence in research and impact (FCAI Director Samuel Kaski)
9:45–10:00: FCAI Special Interest Groups and AIX Forum (FCAI Vice-director Petri Myllymäki)
10:00–11:00: Poster session: FCAI Research Programs and Highlights

FCAI Research Programs and coordinating professors
R1: Agile probabilistic AI, Aki Vehtari
R2: Simulator-based inference, Jukka Corander
R3: Next-generation data-efficient deep learning, Harri Valpola and Alexander Ilin
R4: Privacy-preserving and secure AI, Antti Honkela
R5: Interactive AI, Antti Oulasvirta
R6: Autonomous AI, Ville Kyrki
R7: AI in society, Petri Ylikoski


FCAI Highlights and coordinating professors
A: Easy and privacy-preserving modeling tools, Arto Klami
B: AI-driven health, Pekka Marttinen
C: Intelligent service assistant for people in Finland, Tommi Mikkonen
D: Intelligent urban environment, Kai Puolamäki
E: AI-driven design of materials, Patrick Rinke

FCAI is both a research organization with a sharp research and impact agenda and an open ecosystem of researchers, companies, and public organizations gathered around a common goal. FCAI is creating a new type of artificial intelligence that can operate with humans in the real world – and that can help Finnish industry and society to realize its full AI potential. FCAI is an Academy of Finland flagship, which is a status only granted to six centers of excellence of high quality and high societal impact.


Datanhallinnan ja laskennan eurooppalaiset infrastruktuurit: Suomen näkökulma

Datanhallinnan ja laskennan kehittämisohjelman 3. vuosiseminaari

Ajankohta: Keskiviikko 6.11.2019 klo 12-15

Paikka: Hilton Helsinki Strand, John Stenbergin ranta 4 (Hakaniemi), Helsinki

Anna äänesi kuulua vahvemman osaamisen ja kilpailukykyisemmän datanhallinnan ja laskennan infrastruktuuriympäristön puolesta!

Opetus- ja kulttuuriministeriö toteuttaa tutkimus- ja innovaatiotoimijoiden kanssa datanhallinnan ja laskennan tutkimusinfrastruktuurien, palveluiden ja osaamisen kehittämisohjelman vuosina 2017−2021. Kansallinen kehittämisohjelma linkittyy vahvasti alan eurooppalaiseen tutkimusinfrastruktuuripolitiikkaan. Osaamisen vahvistaminen Suomessa mahdollistaa Euroopan tason laskentakapasiteetin ja palveluiden täysimittaisen hyödyntämisen.

Tänä vuonna DL2021-kehittämisohjelman vuosiseminaarissa tarkastelemme sitä, millaista kansallista ja Euroopan tason datanhallinnan ja laskennan ympäristöä (DL2021, EuroHPC, EOSC) Suomi haluaa olla rakentamassa. Keskustelemme, miten Suomi ja Suomessa työskentelevät tutkijat hyötyvät Euroopan tason palveluista. Selvitämme myös, mitä annettavaa pohjoismaisella yhteistyöllä on Suomelle ja Euroopalle.

Toivotamme seminaariin lämpimästi tervetulleeksi tutkimusorganisaatioiden tutkimusjohdon edustajat ja tutkimushallinnon asiantuntijat, data- ja laskentaintensiivisen tutkimuksen tekijät, ministeriöiden virkamiehet, tutkimuksen rahoittajat sekä tutkimusinfrastruktuureiden ja datan avoimuuden kehittäjät.
 

Tervetuloa!

Erja Heikkinen
DL2021-kehittämisohjelman ohjausryhmän puheenjohtaja
 

Ilmoittautuminen osoitteessa https://www.academyevents.fi/Default.aspx?tabid=929&tap=3721 
viimeistään 22.10.2019

Seminaari on katsottavissa suorana verkkolähetyksenä sekä seminaarin jälkeen tallenteena osoitteessa http://videonet.fi/minedu/20191106/ . Ilmoittautuminen koskee myös etäosallistujia.

Kerrothan tilaisuudesta myös kollegoillesi.

Lisätietoja seminaarista antaa johtava tiedeasiantuntija Anu Nuutinen Suomen Akatemiasta (anu.nuutinen(at)aka.fi; puh. 029 533 5085).


Pouta Cloud + Rahti

* Pouta Cloud Course (13.11.) *

This course gives you a practical introduction for using CSC's Pouta cloud services. The Pouta cloud services are meant for all fields of science & research.
It allows users to solve computational challenges which have not previously been possible to run on CSC's other computing systems. 
Read more and register (by 06.11.) at: https://www.csc.fi/web/training/-/pouta-cloud-course-2019 

* Fundamental of Container Clouds with Rahti (14.11.) *

In this course the fundamental concepts of Kubernetes based container orchestrators are illustrated using the Rahti platform from a practical viewpoint. 
We shall focus on the users point of view for using rahti and aspects important for running and maintaining containerized workloads are emphasized. 
Read more and register (by 07.11.) at: https://www.csc.fi/web/training/-/fundamentals-of-container-clouds-with-rahti 

-- 

OpenID Connect -koulutus
------------------------

8.10. OpenID Connect -protokollan hyödyntäminen käyttäjätunnistamisessa
ja valtuuttamisessa. Rekisteröidy viimeistään: 1.10.2019

Read more >> https://newsletter.csc.fi/go/2855498-275788-15194197


Advanced Gromacs, HADDOCK + PMX Workshop
----------------------------------------

9.-11.10. Two day intensive workshop is organized as a collaboration
with BioExcel, CSC and PRACE. It consists of lectures and hands-on
tutorials introducing HADDOCK, pmx and instructions on how to improve
Gromacs perfomance on HPC platforms. HADDOCK is a versatile
information-driven flexible docking approach for the modelling of
biomolecular complexes. PMX is a service for users who need to do free
energy calculations. Registration deadline: 27.9.2019

Read more >> https://newsletter.csc.fi/go/2855494-271464-15194197


Introduction to Parallel Programming
------------------------------------

23.-25.10. The course introduces the basics of parallel programming with
the message-passing interface (MPI) and OpenMP paradigms. The course
consists of lectures and hands-on exercises. Registration deadline:
16.10.2019

Read more >>
https://newsletter.csc.fi/go/2855493-242942-15194197


Geocomputing in Puhti supercomputer
-----------------------------------

4.11. How to use CSC computing resources for spatial analysis.

methods for spatial data. Registration deadline: 27.10.2019

Read more >> https://newsletter.csc.fi/go/2855499-30063-15194197


Practical machine learning for spatial data
-------------------------------------------

5.-6.11. This course gives a practical introduction to use machine
learning, inc deep learning, methods for spatial data. Registration
deadline: 28.10.2019

Read more >>
https://newsletter.csc.fi/go/2855491-49454-15194197


Schrödinger Maestro Hands-on workshop
-------------------------------------

7.-8.11. Learn to power use the Schrödinger Maestro molecular modeling
suite and the Materials Suite through lectures and hands-on training.
Registration deadline: 30.10.2019

Read more >> https://newsletter.csc.fi/go/2855500-390460-15194197


Pouta Cloud Course
------------------

13.11. This course gives you a practical introduction for using CSC's
Pouta cloud services. The Pouta cloud services are meant for all fields
of science & research. It allows users to solve computational challenges
which have not previously been possible to run on CSC's other computing
systems. Registration deadline: 6.11.2019

Read more >> https://newsletter.csc.fi/go/2855495-301911-15194197


Fundamentals of container clouds with Rahti
-------------------------------------------

14.11. In this course the fundamental concepts of Kubernetes based
container orchestrators are illustrated using the Rahti platform from a
practical viewpoint. We shall focus on the users point of view for using
rahti and aspects important for running and maintaining containerized
workloads are emphasized. Registration deadline: 7.11.2019

Read more >>
https://newsletter.csc.fi/go/2855490-302029-15194197


Practical Machine Learning
--------------------------

19.-20.11. This course gives a practical introduction to machine
learning, including basic approaches to classification, regression,
dimensionality reduction and unsupervised learning. We will cover, among
other things, linear classification and regression, nearest neighbor
methods, support vector machines, decision trees and neural networks.
Registration deadline: 12.11.2019

Read more >> https://newsletter.csc.fi/go/2855496-313635-15194197


CodeRefinery workshop in Espoo
------------------------------

10.-12.12. How is a researcher to know what the best tools and practices
are without investing three to five years in getting a second degree in
computer sciences and engineering? That's what this CodeRefinery
workshop is all about!

Read more >> https://newsletter.csc.fi/go/2855497-424712-15194197


Join our free webinars on data!
-------------------------------

CSC Webinars is a series of short, on-line technical talks followed by a
free-form discussion of the topic at hand or of other issues raised
during the discussion. Everyone is welcome to participate! Join our
webinar series on data below:

15.10. Fairdata PAS-palvelu >> https://newsletter.csc.fi/go/2855492-160073-15194197 Missed a webinar? Don't worry, we record them! You can find all our past
webinars on our YouTube channel https://newsletter.csc.fi/go/2855504-379079-15194197 .

Check all the upcoming courses, webinars and events at https://newsletter.csc.fi/go/2855502-332549-15194197 
--------------------------------------------------------------------------------------------------
* Fortran Programming for Scientific Computing (25.9. - 27.9.) *

The Fortran programming language is still - despite of its long history
- one of the standard and most useful languages in scientific computing.
Our course will give the participants the basic knowledge on the Fortran
programming language. This free of a charge course is part of the PRACE
Training activity. Read more and register (by 18.9.) at:
https://www.csc.fi/web/training/-/fortran-programmin-for-scientific-computing

* Introduction to Parallel Programming (23.10. - 25.10.) *

The course introduces the basics of parallel programming with the
message-passing interface (MPI) and OpenMP paradigms. The course
consists of lectures and hands-on exercises. This free of a charge
course is part of the PRACE Training activity. Read more and register
(by 16.10.) at:
https://www.csc.fi/web/training/-/intro_parallel_program_2019

Webinar: Big Data series: basics of Apache Spark programming

Time: 12.9.2019, 14:00 EET
Lecturer: Apurva Nandan, CSC
Language: English

https://www.csc.fi/web/training/-/big-data-apache-spark

This is the second webinar of the Big Data series - Apache Spark.
Details and materials of the first webinar : https://www.csc.fi/fi/web/training/-/webinar-big-data-spark-2019

This webinar is ideal for people who want to learn more about analysing massive amounts of data. Spark, a popular, high-performance distributed computing framework, allows us to tackle big-data problems by distributing the workload across a cluster of machines, making large scale data analysis rapid and quick!

In this webinar, we will discuss briefly about Spark and learn about the basics of Spark programming in Python!

Prerequisites: As this is the second webinar in the Apache Spark series, we highly recommend that you watch the first part before the webinar: https://www.youtube.com/watch?v=akAwwNxm_HM&feature=youtu.be

Join the webinar in Zoom: https://cscfi.zoom.us/j/717625986

Computing facilities tend to all have the same problems, but every site tends to have its own solutions. Big collaborations tend to be about big hardware or policies, and not so much the day to day difficulties that we face. The Nordics already have a special connection and are a good size to make a grassroots community of admins and support staff working together.

NordicHPC is a group of like-minded individuals in the Nordics who want to work together to make our resources more usable and better maintained (or even just to make the maintenance easier). Despite the name, our focus is not only traditional HPC, but we do tend focus on basic computing infrastructure, not platforms for a specific purpose.

We have a special focus on usability, and a vision to match.

We have a preliminary program scheduled, but the actual program is organized in collaboration with all attendees. We also hope the most important outcome is networking and come out of spontaneous discussions. We’ll use an “unconference”-type format for most of the workshop. Come prepared to collaborate - this is a meeting for the doers.

Registration

You can register here for attendance, possible attendance, or remote attendance.

The ideal attendee is working in developing, maintaining, or supporting general-purpose computing infrastructure in the Nordics, and has an ability to take ideas into practice on their systems. Other users or interested parties are welcome as well. We’ll try to make room for anyone in academic computing centers in the Nordics.

What you need to prepare

  • Think about what cool things you do and what your most common problems are.
  • Prepare to present three cool things and three problems from your site (see below).
  • Suggest a topic or two that we should discuss.

Agenda

We have a preliminary program below, which will continue to be adapted based on attendee interest. We can reserve spaces for any desired break-out sessions or parallel tracks.

  • 14 November.
    • Early arrivals? Come find us in room A243 or the obvious coffee room in the A2 corridor.
    • 11:00 Lunch for those who have arrived
    • 12:00 Afternoon session
      • Talks
      • Breakout sessions, either planned or to continue discussions inspired by talks
    • Evening: group dinner
  • 15 November
    • 09:00 Meeting continues
      • Talks
      • Discussing future the NordicHPC collaboration
      • Breakout sessions
    • Program ends at 12:00, room reserved until 14:00

Contribute!: Request a talk or suggest your own, either in advance or during the meeting, by adding to the agenda google doc. You can also directly contribute by making a Github issue or PR. Slides and material can also be uploaded to the Google Drive. The exact program will be adjusted based on attendee interest closer to time.

Session descriptions

The following sessions have been offered so far:

Cool things and problems: Each site presents around three cool things that their site does, and three problems they have. This becomes a discussion on shared opportunities. Please come prepared with something to present and add to the slides. Organizer: Radovan Bast

Nordic Jupyter: JupyterHub isn’t just a service, but a usable way of accessing services. This workshop shares configurations and knowledge about running JupyterHub for research facilities, and provides hands-on support for sites to set up their own hub. Organizer: Richard Darst

Software deployment: Deploying scientific software is one of biggest pains of sysadmins. Let’s share best practices. We’ll discuss Spack, Easybuild, etc. Aalto University will present its automated build system which automates Spack, Singularity, and more across multiple systems. Organizer: Simo Tuomisto

Lightning talks: Anyone who wants to give a lightning talk, let us know. There’s no topic too small, especially consider talks on small problems, solutions, or UI problems you have.

Requested sessions

If anyone would like to present on these topics, let us know. You don’t have to be an expert, you can also facilitate the discussion. You can also add to the google doc linked above and below.

Managing I/O for machine learning: How to best help users who make millions of small files with a wide variety of machine learning frameworks.

Resources / Sharing

Attending:

  • Meeting video link, coming later
  • Meeting google drive. Anything added here is shared publicly.
  • Meeting presemo (real-time questions and interaction), coming later

Planning:

Location

Aalto University campus (Otaniemi), Finland. Computer Science building room T4. Address: Konemiehentie 2, Espoo.

Arrival and practicalities

This is an unfunded event, so each participant is responsible for their own costs but there is also no registration fee.

Hotels: there is a Radison Blu in Otaniemi which is a common choice. Downtown Helsinki or Espoo hotels are a reasonable choice if close to the metro.

Transportation: from the airport, take the train to the central train station (last stop), then metro to the stop “Aalto University”, then take the B exit. A single ABC ticket (4.60€) covers alltransportation, buy on the train platform at the ticket machines. From Otaniemi to downtown, an AB ticket (2.80€) works. Reittiopas is the journey planner and there is also a HSL app that lets you buy mobile tickets and check routes.

Contact

Richard Darst, richard dot darst at aalto.fi

Python programming
The Python programming language is popular in scientific computing
because of the benefits it offers for fast code development. The
performance of pure Python programs is often suboptimal, but there are
ways to make them faster and more efficient.

On this course, you’ll find out how to identify performance bottlenecks,
perform numerical computations efficiently, and extend Python with
compiled code. You’ll learn various ways to optimise and parallelise
Python programs, particularly in the context of scientific and high
performance computing.

Register now for the first run starting on 9th September 2019 at
https://www.futurelearn.com/courses/python-in-hpc
* Puhti porting and optimization workshop 10.-11.9. *

The focus of the workshop is to facilitate users to port and optimize codes for Puhti. The workshop enables users to get the best possible performance from Puhti. Read more and register (by 30.8.) at https://www.csc.fi/web/training/-/puhti_workshop_june_20-1