Our Team

StreamHPC is one of the best known companies in GPU-computing (OpenCL/CUDA/HIP/SYCL), but we are also very active in embedded development, algorithm-design and technologies like graphics (OpenGL/VULKAN), Machine Learning, and HPC (OpenMP/MPI).

We are distributed between mainly Amsterdam, Budapest and Barcelona.

The developers, the heart of the company

The company consists of highly skilled developers and low-level performance engineers. We mostly manage ourselves, but always with help of the group. This way we have influence by showing ownership.

Each employee regularly shares their experience and checks the work of colleagues, to keep the standards high. This results in faster deliveries with higher quality of code, for which we’ve been complimented often.

Want to work at StreamHPC too? Check our jobs-page.

The Leads

The senior team deals with new directions/markets/strategies, training the employees and making sure the project teams get enabled. We use Holacracy and EOS to lead our company.

HR: Berrak Bas
Consultancy + Projects. Adel Johar
Marketing + Sales: Vincent Hindriksen
General strategy: Vincent Hindriksen
IT: Robin Voetter and Balint Soproni
Operations: Maurizio Campese
Finance: shared
Legal: shared
Open standards: shared
Offices: shared
Products: shared

The shared roles do not need to be filled right now, as they are done together or done outside the company. When these become full-time roles, we will make them vacant and publish them on our jobs-page.

Hire the experts

On average we have our pipeline full for 3-6 months, but always reserve time for shorter projects (maximum a month).

Call +31 854865760 or mail to info@streamhpc.com or fill in the contact form to have a chat on how we can solve your software performance problems or do your software development.

Meet us in April

Posted by Vincent Hindriksen on 4 April 2016

9017503_m The coming month we’re travelling every week. This generates are a lot of opportunities where you can meet the StreamHPC team! For appointments, send an email to contact@streamhpc.com.

Meet us at ParallelCon (6 April 2016, Heidelberg, Germany). Besides the crash course (see below), we also have a talk on Vulkan.
Crash Course OpenCL @ ParallelCon (8 April 2016, Heidelberg, Germany). This is part of the conference – you can still buy tickets!
Meet us in Toronto (11 April 2016, Toronto, Canada). In Toronto for business, with time for appointments.
Meet us at IWOCL (19 April 2016, Vienna, Austria). The event-of-the-year for all OpenCL. So ofcourse we’re there.
Meet us in Grenoble (25 April 2016, Grenoble, France). For a training we’re there the whole week. On Thursday and Friday there is time for appointments.

We’re happy to talk business and about technology. Also giving presentations at your company is an option.

This information was previously communicated via the newsletter and on LinkedIn.

An example of real-world, end-user OpenCL usage

Posted by Vincent Hindriksen on 28 March 2016

We ask all our customers if we could use their story on our webpage. For competition reasons, this is often not possible. The people of CEWE Stiftung & Co. KGaA were so kind to share his experience since he did a OpenCL training with us and we reviewed his code.

Enjoy his story on his experience from the training till now!

This year, the CEWE is planning to implement some program code of the CEWE Photoworld in OpenCL. This software is used for the creation and purchase of photo products such as the CEWE Photobook, CEWE Calendars, greeting cards and other products with an installation base of about 10 million throughout Europe. It is written in Qt and works on Windows, Mac and Linux.

In the next version, CEWE plans to improve the speed of image effects such as the oil painting filter, to become more useful in the world of photo manipulation. Customers like to some imaging effects to improve photo products, to get even more individual results, fix accidentally badly focused photos and so on.

Continue reading “An example of real-world, end-user OpenCL usage” →

Feedback & Privacy

This field is huge and ever-changing, which means that certain old posts might need to get updated information. Also, most of our team is made by humans: a species famous for making all kinds of mistakes.

We care about what you say!

"Feedback is the breakfast of champions."
--Ken Blanchard

We are not native English speakers. Did we say something strange?
Is the site somewhat too technical or is it missing a bit of hard-core code examples?
Is some important information missing?
Is the publishing of the book or Eclipse-plugin taking too long?
Is the site too slow? (or broken in any other way?)
Do you have any compliments? We blush easily!

Tell us what you think by using the contact-page or sending an e-mail to feedback@streamhpc.com.

Privacy

We track the pages you visit with Piwik and Google Analytics, and we use this information to improve our webpage. For Google Analytics there is an opt-out tool, to be excluded from any webpage that uses it. For Piwik you have the choice to opt-out by using the form below. If you opt-out, please be kind to give us feedback on how we can improve our page.

Rapid Performance Assessment

You might have heard about the major speed-ups GPUs and FPGAs have promised, but also about the fact that this speed-up will depend a lot on the type of software/algorithm. Investing in OpenCL or CUDA can therefore feel risky, since going in costs time and money, while keeping out can potentially give too much space to the competition. But if you want your customers to get the best experience without paying an unnecessary high price, you’ll need to know what the return of your investment could be. With this quick assessment we will help you determine exactly that.

What we’ve done before

Most assessments were on answering the question “How much speed-up can I get using GPUs?“. Other questions were:

Does this algorithm work on this specific mobile processor?
Can we better use CUDA, OpenCL or OpenGL shaders for this algorithm?
Does the HPC code run best on a Tesla K40 or FirePro S9150?
How many weeks/months would it take to port all code?
How many GPUs do I need for under 1 second responses?
Does this code port to an FPGA?
Which OpenCL device best suites by algorithm: CPU, GPU, APU, DSP, FPGA or something else?

Is your question in the list?

Program

Within a week we can fully analyse your code, or two weeks if the codebase is large or complex. During the assessment we write/port/optimise code, to be able to support our conclusions with numbers.

After the assessment you get an overview of the hotspots, an indication of total speed-up when using OpenCL (or comparable technology), and the answers to your questions.

Preparations

Send a mail to contact@streamhpc.com for more information, and we’ll call you back to talk about your requirements. Please provide times when you want to be called back.

[button text=”Contact form” url=”https://streamhpc.com/about-us/contact/” color=”red” target=”_self”]

Bits&Chips actie voor OpenCL training

U bent op deze pagina terecht gekomen via de nieuwsbrief van Bits&Chips of op aanraden van een vriend of collega.

Uw doel is om een zware berekening of beeldverwerking te versnellen. Wij leren u dat door u enkele concepten aan te reiken, zodat u uw software op een andere manier ontwerpt en programmeert. Met veel snellere software als resultaat. Dit doen wij aan de hand van OpenCL, een programmeertaal voor parallelle processoren.

De training

In 3½ dag leert u:

Ontwerpen van een parallelle software architectuur,
Gebruik te maken van de grafische kaart als co-processor,
De programmeertaal OpenCL,
Parallelle algoritmes implementeren.

De helft is uitleg door de trainer, de andere helft practica. Op de 4de dag bespreken we een probleem gezamenlijk, om alles samen te laten komen.

Na de training kunt u (zonder speciale tools) langzame code herkennen, deze opnieuw ontwerpen en porteren naar de grafische kaart. U krijgt het lesboek en wat basis-software mee naar huis, zodat u eenvoudig verder kunt oefenen.

De data

De volgende 3 trainingen hebben al enkele aanmeldingen. Ze hebben een speciaal thema en doelgroep.

17-20 maart, Amsterdam. Thema: beeldverwerking.
14-17 april, Amsterdam. Thema: wiskunde/algebra
14-17 juli, Amsterdam. Thema: signaalverwerking

U kunt naar iedere training, als u OpenCL wilt leren. De gebruikte boeken zijn per training verschillend.

De kleine lettertjes

Enkele belangrijke aspecten van de training:

De voertaal is Engels, ivm trainees uit Europa.
Een goede basis-kennis van C is noodzakelijk. U kunt gratis een speciale thuisstudie opgestuurd krijgen.
U dient tijd in te roosteren om na de training te oefenen, zodat u niet snel alles snel weer vergeet.
U heeft een laptop nodig.
Het programmeren gebeurt meestal op een Linux-server. De basis-kennis hiervoor wordt ter plekke uitgelegd.

Aanmelding

De normale kosten voor de training zijn €1850 per persoon. Als u Bits&Chips noemt, krijgt u €100 korting.

Het enige wat u nu hoeft te doen is een mailtje te sturen naar trainings@streamhpc.com met daarin uw gewenste datum en/of onderwerpen. Wij sturen u daarna een vragenlijst terug als voorbereiding op de training.

Voor vragen kunt u terecht bij +31854865760 (kantoor) or +31645400456 (mobiel).

Texas Instruments DSP

TI has a fully conformant OpenCL 1.1 implementation.

The below table is taken from http://downloads.ti.com/mctools/esd/docs/opencl/intro.html and shows which DSPs have OpenCL-support.

SoC	System	Khronos Conformance	Installation Instructions
AM572	AM572 EVM	OpenCL v1.1 Conformant	Processor SDK for AM57x
DRA75x	DRA75x EVM	OpenCL v1.1 Conformant	Processor SDK for DRA7x (Enabling OpenCL on DRA75x)
AM571	AM572 EVM	OpenCL v1.1 Conformant	Processor SDK for AM57x
66AK2H	66AK2H EVM	OpenCL v1.1 Conformant	Processor SDK for K2H
66AK2L	66AK2L EVM	Not submitted for conformance	Processor SDK for K2L
66AK2E	66AK2E EVM	Not submitted for conformance	Processor SDK for K2E
66AK2G	66AK2G EVM	Not submitted for conformance	Processor SDK for K2G

Theoretical Performance of the C66x

Fixed point 16×16 MACs per cycle: 32
Fixed point 32×32 MACs per cycle: 8
Floating point single precision MACs per cycle: 8
Arithmetic floating point operations per cycle: 16 2-way SIMD on .L and .S units (e.g. 8 SP operations for A and B) and 4 SP multiply on one .M unit (e.g 8 SP operations for A and B)
Arithmetic floating point operations per cycle: 164 2-way SIMD on .L and .S units (e.g. 8 SP operations for A and B) and 4 SP multiply on one .M unit (e.g 8 SP operations for A and B)
Load/store width 2 x 64-bit 2 x 64-bit Vector size (SIMD capability): 128-bit (4 x 32-bit, 4 x 16-bit, 4x-8bits)

GFLOPs

2 FLOPs – 2-way SIMD on .L1 (A side) such as DADDSP or DSUBSP
2 FLOPs – 2-way SIMD on .L2 (B side) such as DADDSP or DSUBSP
2 FLOPs – 2-way SIMD on .S1 (A side) such as DADDSP or DSUBSP
2 FLOPs – 2-way SIMD on .S2 (B side) such as DADDSP or DSUBSP
4 FLOPs – 4-way SIMD on .M1 (A side) such as QMPYSP (or CMPYSP, maybe not 4-way SIMD)
4 FLOPs – 4-way SIMD on .M2 (B side) such as QMPYSP (or CMPYSP, maybe not 4-way SIMD)
========================
16 FLOPs total per cycle per C66x CorePac (source)

Boards

A good starter board is the BeagleBoard X-15, and has OpenCL drivers. It has 2x C66X DSPs and 2x 1.5-GHz ARM Cortex-A15.

Request more information about Consultancy

Requesting information does not oblige you to anything.

If you have a quick question, you can best contact us via e-mail, phone or Skype:

e-mail: info@streamhpc.com
phone: +31645400456
skype: vhindriksen or StreamHPC
web form below:

[contact_form]

21-23 August: OpenCL Training London

Posted by Vincent Hindriksen on 13 May 2013

From 21 to 23 August StreamHPC will give a 3-day training in OpenCL. Here you will learn how to develop OpenCL-programs.

A separate ticket for only the first day can be bought, as then will be a crash-course into OpenCL. Module basics.

The second and third day will all about parallel-algorithm design, optimisation and error-handling. Module optimisation with several new subjects added.

The last part of the third day is reserved for special subjects, as requested by the attendees. Continue reading “21-23 August: OpenCL Training London” →

Self-Assessment GPGPU-role

As we’re not a university but a company, there needs to be a balance between things you can offer and the things we offer. Like in every job description, there is a list of bullet points to explain what we seek. To make it possible to self-asses your fitness for the job, we’ve put the number of points (✪) for each bullet point.

INSTRUCTIONS. For each section, assess yourself as being a:

beginner: have been in contact with it briefly
junior: had some experience, but not difficult problems
medior: had more experience, but cannot coach others yet
senior: experienced enough to coach others to really advance on this subject
lead: can teach new things to a senior
principal/master/guru: one of the world’s best

You need to look for the level where you get the most points. For example, if you are a master in C++ but are a medior in C and math, it might actually be best to assess as a medior and mention your C++ knowledge specifically. Or for example, if you are sure you can successfully finish a tutorial on GPGPU, you’re a beginner.

Real question to answer before applying: do you want to become a senior in GPGPU?

If you have the imposter syndrome, don’t be too harsh on yourself. If you’re overconfident, be realistic. If you worry you’re both, pick imposter syndrome only.

Heads up. During the interviews, we ask questions for the above self-assessment. If you assess yourself as a senior for GPU-coding because you were the best-of-class, you’ll get questions. And there is no person who can be defined by lists, so do mention where you stand out.

You, as a CPU Developer (9 items, 18 weights)

We seek people with experience. This can be open source projects for first jobseekers, or past jobs for those with job experience.

You are capable of designing mathematical algorithms, both serial and parallel. ✪
You are strong in math and hard sciences. ✪✪
You know how compilers work, and you are unfortunate enough to know when they don’t. ✪✪
You are experienced in C. ✪✪
You are experienced in C++. ✪✪
You have experience designing performance-driven architectures. ✪✪
You know how to write tests. ✪✪✪
You are experienced with continuous development. ✪✪
You have experience with low-level optimizations. ✪✪

You, as a GPU Developer (6 items, 15 weights)

We seek people with experience. This can be open source projects for first jobseekers, or past jobs for those with job experience.

You have read “hardware architecture specification documents” or ISA-docs. ✪
You know how GPU-compilers work, and you are unfortunate enough to know when they don’t. ✪✪
You are experienced in CUDA and/or HIP. ✪✪✪✪
You are experienced in OpenCL and/or SYCL. ✪✪✪✪
You know your way around with GPU-libraries. ✪
You are experienced in porting algorithms to the GPUs, without the use of any library. ✪✪✪

As the four stars indicate, we do need minimal GPGPU-experience, as you won’t learn it here.

You, as a Problem Solver (8 items, 21 weights)

Coding is only one part of the solution. Most of the time we’re solving problems, where coding is just the means.

You like the ideas and theories around the “learning mindset”. ✪✪✪
You have a structured problem-solving approach that you could explain. ✪✪✪
You have high self-awareness and can self-observe. ✪✪✪✪
You have high standards for yourself. ✪✪
You test out approaches by making quick experiments. ✪✪
You test out possible solutions by mentally putting them in different scenarios. ✪
You regularly take time to zoom out to get an overview on the problem, to be able to balance the inputs for the solution. ✪✪✪✪
You always follow through. ✪✪

If you score high here, this will compensate for any lack of technical experience. Also for continuous growth, you’ll need to score high here.

You, as a Project Team Member (10 items, 20 weights)

Our company’s strength is that we work in teams. We don’t know everything as individuals, but as a team we can solve almost any problem around HPC and GPUs. This means we highly value collaboration and thus must be efficient in project handling.

You have a proven track record of being focused on results. ✪
You have a talent for turning vague problems into the right actions, and you want to build on it. ✪
You normally write down tasks, and then prioritize & ESTIMATE them. ✪✪✪
You understand that well-defined, well-communicated delivery criteria are the responsibility of every team member. ✪✪✪
You can identify something’s missing to move a project forward smoothly. ✪✪✪
You speak up when the project diverges from the trajectory. ✪✪
You are used to administrating your time spent on an issue. ✪
You can delegate work. ✪✪
You can get work delegated. ✪✪
You can explain, with examples, why the above are important. ✪✪

We explicitly did not state “project management”. It is about playing your part of making an efficient team.

What’s next?

If you got at least junior on CPU, Team, and problem-solving, beginner on GPU, and at least one of the 4 on medior? Then you should apply. Go to https://streamhpc.com/jobs/ for the instructions and links to other articles that should help you with understanding if this is a job for you.

Understand that if you are a true beginner in GPGPU, it’s best to follow the tips&tricks explained here.

Win an OpenCL mug!

Posted by Vincent Hindriksen on 18 March 2017

The first batch is in and you can win one from the second batch!

We’re sending a mug to a random person who subscribes to out newsletter before the end of 17 April 2017 (Central European Time). Yes, that’s a Monday.

Two winners

We’ll pick two winners: one from academia and one from industry. If you select “other” as your background, then share which category you fall in the last field.

Did you already subscribe and also want to win? I am not forgetting you – more details are in a newsletter next quarter.

More winners, by referring to a friend

If you refer a colleague, a friend or even a stranger to subscribe, you can both win a mug. Just be sure he/she remembers to mention you to me when I ask. Before you ask: the maximum referral-length is 5 (so referral of referral of referral of referral, etc) plus the one who started it.

UPDATE: If you win a mug and were not referred by somebody, you can pick a co-winner yourself. Joy should be shared.

Newsletter

You can also use this link http://eepurl.com/bgmWaP.

Code Review

Code reviews are one of the fastest ways to get the dev-team back on track in order to add performance to the code. We offer two types of code reviews, all safely under an NDA. This way you keep in control of the development, while getting expert-knowledge in.

A quick scan gives you an overview of the main ways to speed up the code and how it can be done.
This quick scan can be delivered in one week, if necessary, to give you the direction you may require in times of pressure.

Also, an extensive code review can provide all the necessary information for a redesigned architecture.

GPU-code (OpenCL, CUDA, Aparapi, and more)

Writing GPU-code and performing host-code can be tricky. The best method to learn CUDA or openCL is by doing. Nevertheless, you may need feedback sometimes to be sure you’re doing the right thing. We can check your code and give you a report with hand-on tricks to make it optimal.

CPU-code (Java, C, C++ and more)

Many CPU-codes, like Java, C, C++ and C# are written with functionality in mind, but not performance. Adding performance (cache-optimisation, memory-usage reduction, parallelisation of computations, adding OpenMP-threads, etc) is quite doable, but only when you know how. We can help you increase performance of the software through feedback and clear steps.

Let us help you!

If you are interested in this service, request more information today and we will get back to you as soon as possible. Of course, you can also contact us via phone (+31 6 45400456), or e-mail (info@streamhpc.com).

Events&Talks

StreamHPC gives talks at public and in-company events to explain what GPU-programming is, while focusing on the day’s theme.

You are welcome to attend these days, or you can request a talk about OpenCL and GPU-programming to be given at your event.

Agenda Talks

At the events in the list below Vincent Hindriksen will give a talk, or has given a talk.

Date	Location - Language	Description	Link to program	Type
6 November 2013	Nijkerk - English	Aparapi and Project Sumatra: using GPGPU in Java	NLJUG J-Fall 2013	Registration and NLJUG-membership required
31+31 October 2013	Cambridge (UK) - English	Using the GPU for Physics computations via OpenCL	Mosaic3DX	Registration required
4 October 2012	Amsterdam - English	Introduction to OpenCL on mobile processors, to make hackers&funders think of new ideas for products which were never possible before.	Hackers and Founders (Amsterdam, NL) Meetup	Free
20 September 2012	English	~~Ongoing work on the OpenCL plugin for Eclipse - presented remotely.~~ I cancelled this, because my work has not advanced enough.	PTP User-Developer Workshop Sept 18-20, Chicago	Registration required.
28 June 2012	Amsterdam - English	GPGPU-day organised by StreamHPC. No talks by us, but by many interesting speakers from many Dutch universities.	Platform Parallel NL	Registration required. Free for students and researchers in the Netherlands.
20 June 2012	Delft - English	Industry-session at HPDC'12 (session 4). "Parallel Programming for the Masses" about how to walk the road forward while on the road of legacy.	HPDC'12	Registration required.
15 June 2012	Ede - Dutch	Talk at SDN about how to use GPU-programming in .NET, including introduction to GPU-programming.	SDN	Registration required. Free for SDN-members.
25 May 2012	Amsterdam - English	In-company talk about parallel and GPU programming to decrease power usage.		Internal

Reservations

For reservations and requests, please mail to events@streamhpc.com.

Agenda Events

We are visiting or have visited the following events. This is perfect if you want to have a quick discussion with us.

Date	Location	Description	Link to program
20-22 January 2014	Vienna, Austria	A premier forum for experts in computer architecture, programming models, compilers and operating systems for embedded and general-purpose systems.	HiPEAC
12+13 May 2014	Bristol, UK	An annual meeting of OpenCL users, researchers, developers and suppliers to share OpenCL best practise, and to promote the evolution and advancement of the OpenCL standard.	IWOCL
20 June 2013	Amsterdam, Netherlands	All about GPGPU in the Netherlands	Applied GPGPU-days
21+22 May 2013	London, UK	All about HPC-techniques on low-power devices	LEAP-conference
26-28 February 2013	Nürnberg, Germany	873 exhibitors from 37 countries. Will focus on the processors with high-end compute-capabilities.	Embedded World'13 Exhibition
21-23 January 2013	Berlin, Germany	International conference on high-performance and embedded architectures and compilers.	HiPEAC'13
18 December 2012	Paris, France	Cancelled. ~~Meetup by Paris HPC group. This talk will be about most efficient (GFLOPS / Watt) processors existing today.~~	Very High Performance/Watt Processors: the Road to Exascale
17 December 2012	Amsterdam, Netherlands	The e-BioGrid project is part of the BiG Grid project to establish an e-infrastructure for life sciences.	e-BioGrid and beyond
13 December 2012	Brussels, Belgium	All around GPUs, FPGAs and upcoming architectures.	Symposium on Personal High-Performance Computing
5 + 6 November 2012	Amsterdam, Netherlands	Big data event around Smart Systems, Cloud Computing, Mobile and Social Media. (I did a pitch talk here)	Perfect Storm Europe
29 November 2012	Eindhoven, Netherlands	Altera explains how FPGAs with an ARM-core can work for many types of problems.	Altera Soc FPGA event Altera SoC FPGA event
24 - 26 October 2012	Eindhoven, Netherlands	Talks around multiscale science.	Opening Symposium Eindhoven Multiscale Institute
26 September 2012	Amsterdam, Netherlands	Alles over Grid-computing in Nederland.	BiG Grid and beyond

Dutch: Gratis kennisochtend over de nieuwe generatie processoren

Posted by Vincent Hindriksen on 26 September 2011

Ergens opgevangen dat grafische kaarten tegenwoordig ingezet kunnen worden voor zware berekeningen? Tijdens een koffiegesprek gehoord over vector-processors als aanvulling op scalaire processors? Dan wordt het tijd dat u de grote veranderingen op processorgebied op een rijtje krijgt om uw organisatie beter op innovatief gebied te kunnen sturen.

Zie https://streamhpc.com/education/gratis-kennislunch/ voor een uur uitleg op lokatie.

Voor wie is deze kennis-ochtend?

Bedrijven voor wie snelheid belangrijk is en grote hoeveelheden data moeten verwerken. Bijvoorbeeld rekencentra, R&D-afdelingen, financiele instituten, ontwikkelaars van medische software, algoritme-ontwikkelaars en vision-bedrijven. Ook investeerder met hitech-bedrijven in hun portfolio kunnen gratis op de hoogte gebracht worden van de huidige ontwikkelingen.

U heeft geen technische achtergrond nodig, maar u zult zich niet vervelen indien u bits&bytes spreekt. Wij vragen uw achtergrond aan te geven, zodat we de juiste details in het programma kunnen toevoegen.

Wat is het programma?

In het eerste uur hoort u hoe de huidige processor-markt veranderd zijn ten opzichte van enkele jaren geleden – en welke nieuwe software-ontwikkelmethodes zijn geintroduceerd. Daarna krijgt u een overzicht van de nieuwe oplossingen die beschikbaar zijn en hoe dit zich verhouden tot de bestaande. Dit geeft u dan voldoende inzichten om te bepalen of het toepasbaar is binnen uw bedrijf. Het uur wordt afgesloten met wat StreamHPC voor u kan betekenen, maar ook wat u zelf kunt doen.

In het tweede uur bespreken we enkele use-cases en is er tijd voor vragen. De use-cases die worden besproken zijn afhankelijk van de achtergronden van de aanwezigen; denk aan bijvoorbeeld Monte Carlo, physics, enzym-werkingen, matrix-berekeningen en neurale netwerken.

Wanneer?

Indien er minimaal 10 aanmeldingen zijn, wordt er een datum geprikt.

Indien u binnen uw bedrijf direct interesse heeft, is het mogelijk dat StreamHPC bij u langs komt om deze presentatie te geven aangepast aan uw achtergrond. Neem daarvoor contact met ons op.

Online event registration for De opkomst van de supersnelle processoren – wat betekent dit voor u? powered by Eventbrite

Help us find our future COO

Posted by Vincent Hindriksen on 20 July 2018

*Is this a motto that goes with your personality? Then we want to talk with you.*

About 7 years ago we were still dealing with the usual peaks and lows of consultancy.

I’d like to get your help to find our future COO to help streamline this growth.

You might have seen that there are hardly any new blog posts – now you know why. By helping us find that special person, there can be put more time of writing new blog posts again.

If you know the perfect person for this job in Amsterdam, please let them know there is this unique company looking for her or him. Sharing this blog-post would help a lot.

You can find more information in this job-post:

We all know that quality comes with attention to detail, but also that with growth the details are the first to be postponed. We seek help in handling daily operations during our growth. The most important tasks are:

Customer contact. You make sure the communication is regular and smooth with all our customers, making them more engaged and happy with us.

Sales follow up. You take over to discuss the needs of potential customers pre-sales has had contact with.

Team support. You help the development-teams to get even better by helping them to solve their daily and long-term problems.

The job is very broad, but is all around a listening ear and getting things done.

You have studied business administration or alike, and have a can-do attitude. You know how to work with technical people and are a real team-player. You understand how to develop and engage group dynamics.

Do you think this is a job written for you, then we would like to hear more from you! Send an email to jobs@streamhpc.com with a motivational letter and listing relevant experience.

Thanks for helping out!

If you got sent here, we hope to hear from you!

About Us

Stream HPC is a software development company in parallel software for many-core processors. We provide professional software development services, training and consulting to help you increase compute performance in software while lowering hardware-costs.

We have 3 locations.

Stream HPC B.V. (Amsterdam)

Koningin Wilhelminaplein 1 – 40601
1062 HG Amsterdam
Netherlands, Europe

phone: +31 854865760 (office) or +31 6 45400456 (cell)

Visit us in Amsterdam

Stream HPC Hungary Kft. (Budapest)

Radnóti Miklós u. 2
Budapest, 1137
Hungary, Europe

Stream HPC Spain S.L. (Barcelona)

Plaza de Catalunya 1, 4th floor
Barcelona 08002
Spain, Europe

History

2010 – 2013: the freelancing years

The company started as a freelancing business, with one focus: Programming GPUs with OpenCL. It was though, as back then the G in GPU stood for “Graphics only”.

The name was “StreamComputing” = A high-performance computer system that analyzes multiple data streams from many sources live. The main goal was to create software algorithms that analyze the data in real time as it streams in to increase speed and accuracy when dealing with data handling and analysis, which was in line with that.

2014: first hope

Four years later the first employee, Anca, was hired. Later that year the freelancing business was was turned into a limited company. GPUs got more seen as data-processors and trainings were the main income. Projects were still small, GPGPU was a world of early adopters and most time was invested on trainings.

First contact was made with AMD, now one of our biggest clients.

2015-2017: initial growth

Stream grew to a handful of employees, and we did projects for HSA foundation, Stanford, AMD, Zeiss, Nokia, Philips and many lesser known companies.

Trainings were still done, but were by far not the main resource of income anymore. We tried some FPGA-work, but found that most promises were not implemented yet.

2017: a new name

We renamed the company to Stream HPC. There were several reasons. As we focused more on customers from Asia and North America, we needed the .com, which was unavailable. Getting the new name was quite a quest, but we got there: by customers we were often referred to as “Stream”, a business coach assured us that CPU-work would remain important and thus “HPC” was more important that “GPU”, and it was quite difficult to type streamcomputign correctly.

2017-2020: hitting all kinds of ceilings

The goal was to grow further, but this turned out to be more difficult than expected. All kinds of obstacles got in our way, and we even once shrunk in size. With trainings, coaching, reading and persistence, we got to understand the hurdles and finally could implement solutions. Looking back it was easy.

2021: Stream HPC Hungary

Hungary started as a group of freelancers. We were very happy with the quality provided by our Hungarian colleagues, and that was enough reason to invest more. We opened the new office in Q3.

The company now turned into a group of companies, and all was set up to extend the group more easily.

We grew back to 15 people by the end of the year.

2022: Benchmark.io

At ISC Benchmark.io was started. To help our customers do better benchmarks, we put all our knowledge into a separate product. Due to high demand for our consultancy services, it is in private beta only.

2022: Stream HPC Spain

Barcelona was opened in Q3.

The estimation is to grow to 25-35 people by the end of the year.

Slow Software Hotline

Posted by Vincent Hindriksen on 15 June 2017

In the perfect world all software is fast, giving us time to do actual work. Unfortunately we live in an unperfect world, and we have to spend extra time controlling our anger as the software keeps us waiting.

Therefore we have opened the Slow Software Hotline – reachable via both phone and email. It has one goal: make you feel happy again.

Reporting is easy. Just name the commercial software that needs to be sped up and why. We’ll do the rest. If you need help with initial anger management due to the slow (and/or buggy) software, we’re happy to help with breathing practises.

Phone: +31 854865760

Email: hotline@streamhpc.com

We will not sit still until all software is fast. Speeding up all software out there. One at the time.

Should SPIRV be supported in CUDA?

Posted by Vincent Hindriksen on 7 March 2017 with 6 Comments

Would you like to run CUDA-kernels on the OpenCL framework? Or Python or Rust? SPIRV is the answer! Where source-to-source translations had several limitations, SPIRV 1.1 even supports higher level languages like C++.

SPIRV is the strength of OpenCL and it will only get bigger.

Currently Intel drivers best supports SPIRV, making it the first target for the new SPIRV-frontends. It is unknown which vendor will be next – probably one which (almost) has OpenCL 2.0 drivers already, such as AMD, ARM, Qualcomm or even NVidia.

How interesting is SPIRV really?

So if SPIRV is really that important a reason to choose for OpenCL, I thought of framing it towards CUDA-devs on twitter in a special way:

Should CUDA 9 support SPIRV?

So that’s a big yes for SPIRV-support on CUDA. And why not? Don’t we want to program in our own language and not be forced to use C or C++? SPIRV makes it possible to quickly add support in any language out there without official support of the vendor. Let wrappers handle the differences per vendor and let SPIRV be the shared language for GPU-kernels. Where do we need OpenCL or CUDA for, if the real work is defined by SPIRV?

What do you think? Leave your comment below how you see the future of GPGPU with SPIRV in town. Is 2017 the year of SPIRV?

And if you worked on a SPIRV-frontend, get in touch to continue your project on https://github.com/spirv. Yes, it’s empty right now, but you don’t know what’s hidden.

Winning demo of Tokyo Demo Fest 2013 uses OpenCL

Posted by Vincent Hindriksen on 15 May 2013

The Tokyo Demo Fest 2013 is one of the many demo-parties around the globe. At such parties is where great programmers meet great artists and show off what came out of their collaborations.

The winner of this year used OpenCL to compute real-time procedurally generated geometries. For the rest C++, OpenGL and Gamemonkey Script was used.

http://vimeo.com/59398253

Tech features: curl noise, volumetric textures, Perlin noise, mesh deformations, HDR/bloom, film grain, fractals, Hermite splines, Tweens and quaternion iridescent particles.

The creator, Tokyo-based Eddie Lee, has done more projects – be sure to visit his homepage. I hope more demosceners start using the power of OpenCL to get more out of their demo’s.

Do you see where below kernel is used? Hint: check the other videos of Eddie.

__kernel void curlnel( 
                      __read_only image3d_t volume,  
                      sampler_t volumeSampler,  
                      __global float4 *output,  
                      float speed 
                      ) 
{ 
    uint index = get_global_id(0); 
    uint numParticles = get_global_size(0); 
    float indexNormalized = (float)index/numParticles; 

    // read from 3D texture 
    float4 vertData = output[index]; 

    float3 samplePos = vertData.s012; 
    samplePos = samplePos+(float3)(0.5f); 

    float4 voxel = (read_imagef(volume, volumeSampler, 
                   (float4)(samplePos,1.0f))); 

    vertData.s012 += voxel.xyz*speed; 

    output[index] = vertData; 
}

According to GPUVerify (see previous post) the line starting with “float4 voxel” has an error.

StreamHPC flirts with ARM

Posted by Vincent Hindriksen on 8 March 2012 with 2 Comments

With the launch of twitter-channel @OpenCLonARM we now officially show a strong interest in ARM for compute. And we are not the only ones, as the twitter already has 80 followers (60 in 1.5 day and 12 retweets of the welcome-message).

ARM has made tremendous progress in both technology and market-share. With ARM-64, companies like NVidia (and maybe AMD) in the field, X86 seems to be getting a real competitor. This could happen because since a few years computers are fast enough and are not being replaced by a faster one, but a smaller one (tablet, phone) or extra one. By the rules of the market, current technologies are replaced by the ones that give those other needs. ARM is fast (enough), flexible in design, very cheap, low-power and passively cooled. The biggest obstacle seems to be only getting a standard for a docking-station to connect your mobile, tablet or watch to keyboard, mouse and large screen.

OpenCL is perfect for ARM, as it gives the computation-power to the intensive computations not already covered by hardware-support. In the world of X86 this interests high performance and big data companies, where on ARM this interests also more. Without the need for OpenCL you can already watch HD video, with OpenCL you can encode the video with MP4. This year you will certainly hear more about new possibilities of OpenCL on ARM.

What do you think. Why does Intel not sell IP to ARM-companies as many technologies could be reused? Could Intel be the next ARM as an IP-seller, or will they stay the defender of X86 for many years to come?

streamhpc.com is not affiliated with ARM.

New grown-ups on the block

Posted by Vincent Hindriksen on 24 October 2010 with 2 Comments

There is one big reason StreamHPC chose for OpenCL and that is (future) hardware-support. I talked about NVIDIA versus AMD a lot, but knowing others would join soon. AMD is correct when they say the future is fusion: hybrid computing with a single chip holding both CPU- and GPU-cores, sharing the same memory and interconnected at high speed. Merging the technologies would also give possible much higher bandwidths to memory for the CPU. Let us see in short which products from experienced companies will appear on the OpenCL-stage.

Continue reading “New grown-ups on the block” →