Vacature: High performance computing

Do you like to be challenged and work independent in an innovative environment. Do you want to be able to reach out to the expertise of more than 130 smart colleagues, experienced in high-tech and state of the art development from edge-to-cloud? Do you like to join various Tech and Fun events like weekly Friday afternoon drinks, TechTalks, sports activities and celebrations? And most important; do you want to be able to develop and grow as a professional?   

Then join us at TOPIC! 

“Graduating at TOPIC has been a very fun and educational experience for me. The complex challenges I have encountered during my graduation internship have provided me with lots of great technical and non-technical experience that I will forever take with me during my career.” 

Mike Huisman | Hardware Engineer at TOPIC

Your role as graduate student 

TOPIC develops embedded systems that are essential for products manufactured by renowned high-tech companies around the world.  Our projects involve edge to cloud solutions; from PCB design and FPGA to low level embedded software and application software for UI, Web, Mobile and cloud. We like to apply innovative technologies such as object tracking, Ultra-Wideband, Lidar, generating/rendering 2D to 3D medical images and Photonics.    

As a graduate student at TOPIC you will have the opportunity to work on innovative R&D subjects that will bring value to our organization. There are always some assignments in mind, but we are also open for your suggestions and can customize them to a suitable assignment. 

About the assignment 

For the more challenging projects, the computational power of systems can become a bottle neck. This can be for block-chain algorithms, signal processing, digital imaging (MRI, 3D projections/overlay), Radar, Lidar, photonics, ML/AI algorithms, etc. The obvious solution is the use of GPU cards, like the ones from Nvidia. But most of the time an FPGA is a better choice to match the algorithm architecture with the implementation structure. 

TOPIC has been using Alveo U50 FPGA cards for a specific application. These are FPGA devices that incorporate PCIe GEN4 interfaces for seamless integration with PC systems, on-board high-bandwidth memory (HBM) for very rapid and massive off-loading of temporary data streams, thousands of DSP cores and distributed memory block as well as loads of random logic. There is a basic PCIe implementation available, including a Linux driver. 

The assignment is to bring-up a HPC system, that consists of 1 PC running Linux, incorporating 4 Alveo U50 boards. All boards should run the same FPGA image and collaborate in the context of video processing of full-HD or 4K video at 60 fps. The HBM memories can be used for temporary video storage and the boards can also interact with each other using an SFP port. The same SFP port can be used to stream video content to our video wall setup to visualize the processing results. 

The assignment will involve Linux based application development, FPGA design and implementation, (optionally) the use of High-Level Synthesis to transform untimed C-models into VHDL code and a communication link. The assignment works best when working with two students on this project due to complexity and amount of work. 

As a graduate student at TOPIC 

• You are graduating from University or HBO in Electronics, (applied) Computer Science, Computer Engineering, Electrical Engineering, Embedded Systems, Physics, Mathematics or similar field; 
• You are pro-active, able to think outside the box, a problem solver and are capable of communicating with and retrieving information from different stakeholders. 
• You have the ability to self-steer, a drive to deliver but also like to have fun while reaching your goals!