This PATC will be offered f2f in Severo Ochoa Room, Capella and videoconferenced online via Zoom. Please, when registering, select how do you plan to particpate.
Please, bring your own laptop. All the PATC courses at BSC are free of charge.
Course convener: Rosa Badia, Workflows and Distributed Computing Group Manager, Computer Sciences - Workflows and Distributed Computing Department
Lecturers:
Rosa M Badia, Workflows and Distributed Computing Group Manager, Computer Sciences - Workflows and Distributed Computing Department, BSC
Javier Conejero, Senior Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC
Jorge Ejarque, Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC
Daniele Lezzi, Senior Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC
Francesc Lordan, Postdoctoral Researcher, Computer Sciences - Workflows and Distributed Computing Department, BSC
Objectives: The objective of this course is to give an overview of the COMPSs programming model, which is able to exploit the inherent concurrency of sequential applications and execute them in a transparent manner to the application developer in distributed computing platform. This is achieved by annotating part of the code as tasks, and building at execution a task-dependence graph based on the actual data used consumed/produced by the tasks. The COMPSs runtime is able to schedule the tasks in the computing nodes and take into account facts like data locality and the different nature of the computing nodes in case of heterogeneous platforms. Additionally, recently COMPSs has been enhanced with the possibility of coordinating Web Services as part of the applications. COMPSs supports Java, C/C++ and Python as programming languages.
Learning Outcomes: In the course, the COMPSs syntax, programming methodology and an overview of the runtime internals will be given. The attendees will get a first lesson about programming with COMPSs that will enable them to start programming with this framework.
A hands-on with simple introductory exercises will be also performed. The students who finish this course will be able to develop simple COMPSs applications and to run them both in a local resource and in a distributed platform (initially in a private cloud). The exercises will be delivered in Python and Java. In case of Python, Jupyter notebooks will be used in some of the exercises.
Level: for trainees with some theoretical and practical knowledge.
INTERMEDIATE: for trainees with some theoretical and practical knowledge; those who finished the beginners course
ADVANCED: for trainees able to work independently and requiring guidance for solving complex problems
Prerequisites: Programming skills in Java and Python
Tentative Agenda:
Day 1 (January 25th, 2022)
9:30 - 10:00 - Welcome and round table (30')
10:00 - 10:30 - Session 1: Introduction to COMPSs (30')
10:30 - 11:15 - Session 2: PyCOMPSs: Writing Python applications (45')
11:15 - 11:45 - Coffee break
11:45 - 13:00 - Session 3: Python Hands-on using Jupyter notebooks (1h15')
13:00 - 14:30 - Lunch break
14:30 - 15:00 - Session 4: Machine learning with dislib (30')
15:00 - 16:30 - Session 5: Hands-on with dislib (1h30')
Day 2 (January 26th, 2022)
9:30 - 11:00 - Session 6: Java & C++ (1h30')
11:00 - 11:30 - Cofee break
11:30 - 13:00 - Session 7: COMPSs Advanced Features (1h30')
13:00 - 14:30 - Lunch break
14:30 - 16:30 - Session 8: Cluster Hands-on (MareNostrum) (2h)
16:30 - 16:45 - Session 9: COMPSs Installation & Final Notes (15')
END of COURSE
