3.5 day OpenCL training in Amsterdam (media)

#_TOWN, 17 Mar 2014

Every month StreamHPC offers an OpenCL training in Mathematics or Media-operations. Target is OpenCL 1.2 (or 1.1 when NVIDIA is discussed). OpenCL 2.0 trainings will start in Q2/Q3, or when on request. All trainings will be given by experienced OpenCL developers/trainers.

Trainings

A trainings takes 3½ days, from basics on the first morning to the special requests on the 4th morning. You’ll bring your own laptop, or you can login to a compute-server.

The Media-operations module is based on “Heterogeneous Computing with OpenCL, second edition” by Benedict Gaster, Lee Howes, David R. Kaeli, Perhaad Mistry & Dana Schaa.

It covers convolution, video-processing, histogram and mixed particle simulation. Extra subjects are OpenCL-OpenGL interop and code-optimisation.

It is fit if you work with images, sound and video.

 

Both training-modules are based on a book, and enriched with nowhere-to-be-found information, explanations about parallel algorithm design and the examples+experience of the instructor, Vincent Hindriksen. About 50% of the training is lectures and 50% practicals. The training materials get improved every time

Hardware Architectures

General optimisations will be discussed, focusing on NVIDIA (limited to OpenCL 1.1), AMD, Intel, ARM MALI, Imagination PowerVR, Qualcomm Adreno, Freescale/Vivante. Specific hardware-targets are by request.

Location

Default location is Amsterdam, where this training also will take place. It has a regular and cheap flight-connection to all mayor cities in Europe, good public transport and relatively good hotels and hostels.

Alternative locations for upcoming events can be chosen, if at least three weeks upfront. Locations can be, but not restricted to:

  • Berlin
  • London
  • Paris
  • Barcelona
  • Dublin
  • Vienna
  • Geneva
  • Stuttgart
  • Brussels

Costs

The costs are €1850,- per person (including a copy of the book), when at least 5 people have subscribed to the course. Academic discounts available.

If you want a private, in-house trainings, please contact us for more information. Trainings outside Europe (Russia, Asia, Americas) are possible, but travel-costs will be added when the group is under 6 people.

Schedule your training

Apply by using the contact form and by filling in the pre-training questionnaire.

We try to get groups together which have common interests. To give you a better experience, you can yourself be rescheduled to another group that is more fit.

If you have any question, just ask. Call +31854865760 (office) or +31645400456 (cell), or use the above contact-form.

See you in class! You will soon be making faster software.



Want to know more? Get in contact!

We are the acknowledged experts in OpenCL, CUDA and performance optimization for CPUs and GPUs. We proudly boast a portfolio of satisfied customers worldwide, and can also help you build high performance software. E-mail us today

#_MAP