Birthday present! Free 1-day Online GPGPU crash course: CUDA / HIP / OpenCL

Stream HPC is 10 years old on 1 April 2020. Therefore we offer our one day GPGPU crash course for free that whole month.

Now Corona (and fear for it) spreads, we had to rethink how to celebrate 10 years. So while there were different plans, we simply had to adapt to the market and world dynamics.

5 years ago…

Setup of the training

Getting your mind tuned from CPU-focused to GPU-focused is half the work. This is the first goal of the day.

The second goal is hands on. We’ll provide templates for CUDA/HIP and OpenCL, where you focus on making the core compute part.

The third goal is learning from a Q&A. Here you can tap into 10 years of our experience with HPC and GPUs, as we did the strangest to the most normal projects.

The training is very basic, but there will be a lot of aha!s that will help in making better decision for adding GPU-support to your project.

This is the agenda:

  • Generic GPGPU model
  • OpenCL language
  • CUDA/HIP language
  • Memory objects
  • General hardware overview
  • Task-parallelism and data-parallelism
  • Mapping code to CPUs and GPUs
  • Comparison to other languages like SYCL, OpenMP and OpenACC

Inbetween there will be short lab-sessions on your own laptop.

Q&A

  • Q: Will I be able to write a simple GPU program after one day?
  • A: Yes, we’ll share free templates of CUDA, HIP and OpenCL you can use
  • Q: I’m a very senior developer. Will I be a senior in GPU-development after the training?
  • A: No. It depends on the person, but it can take multiple years before you reach proficiency in GPU-development. This training mostly helps in avoiding some time-consuming paths
  • Q: I’m in Asia/America. Will you hold a training in my timezone?
  • A: no and yes. It will be two half days.
  • Q: Why not SYCL, OpenMP or OpenACC?
  • A: Not this time. We certainly see a rising demand, but the material is not finalised enough. We therefore keep it to a comparison.
  • Q: How good should my internet be?
  • A: Fast enough for live video.
  • Q: Do I need to have set up an environment on my own computer?
  • A: Yes. GPU-drivers, CUDA, HIP or OpenCL, Cmake and Git. You’ll get instructions after admission.
  • Q: Do I need background in programming.
  • A: No. But be sure that a colleague who can program also joins.
  • Q: Do I need good understanding of C and C++?
  • A: You need to understand pointers, to not delay the programme. If you don’t, then you need to prepare some material before the training. This material is given to everybody. During the training pointer-questions will not be answered.
  • Q: When exactly is it?
  • A: Share the dates that suit you, and we’ll use schedule-polls to find a date that suits everybody. The more participants, the more choice.

More information you possibly will find in the description of the full training. Especially if you want to have a full in-company training, that page tells much more.

Admission

Copy the below template and send to info@streamhpc.com. A personal text is appreciated, ofcourse.


Hi Stream Team,

Congrats with 10 years.

  • Name:
  • Company/institution:
  • Primary programming languages:
  • Reason for interest:
  • Timezone:
  • Preferences on dates:
  • Optional: important extra info:

Kind regards,


We always contact you within 5 days, but we try the same day.

The exact way of organisation depends on the number of interested people and their timezone. As the training is not designed as a MOOC, we can limit the number of participants – first come, first served. We’ll share updates on Twitter, on LinkedIn and in updates in this blog-post.

Looking forward to virtually meet you!

Vincent Hindriksen and the Stream Team

Related Posts

FIA_F1_Austria_2018_Nr._33_Verstappen

The Art of Benchmarking

How fast is your software? The simpler the software setup, the easier to answer this question. The more complex the software, the more the answer will ...

network-of-boxes

Problem solving tactic: making black boxes smaller

We are a problem solving company first, specialised in HPC - building software close to the processor. The more projects we finish, the more it's clea ...

stocks

Improving FinanceBench

...  and preliminary results. Work done before Ofcourse the first step was selecting good algorithms that had both needed a lot of ...

Kabul - An Afghan National Police (ANP) students study basic logistics training at the Afghan Ministry in Interior (MOI) August 1, 2010. The Afghan MOI is providing the first ever computer based logistics training for ANP personnel. (U.S. Air Force photo/ Staff Sergeant Matt Davis)

Updated: OpenCL and CUDA programming training – now online

...  trainings. It includes: Four days of training onlineFree code-review after the training, to get feedback on what you created ...