Development of a heterogeneous computing platform based on IBM’s DOME MicroDataCenter and cloudPFGA high-performance computing platforms
Javier
Last updated
As CTO & Head of Product at DC-ONE AG (later called GRAPIN AG), Javier was in charge of developing a cutting-edge technology platform for heterogeneous computing based on IBM’s Research Europe Zürich DOME MicroDataCenter and cloudFPGA high-performance computing platforms.
IBM’s DOME MicroDataCenter
1.22 MB
View full-sizeDownload As an FPGA expert, Javier was in charge of holding technical discussions with IBM’s cloudFPGA team lead for scoping the platform redesign to bring IBM’s prototype to be production-ready. As a product designer, Javier worked together with IBM’s cloudFPGA Department Head (Cloud and Computing Infrastructure) and Business Development Executives to deal with all sort of business and collaboration aspects. Additionally, Javier was involved in the following activities:
Development of a proprietary platform for heterogeneous computing based on modern single-source SYCL/DPC++ programming models and Jupyter notebooks,
Development of an FPGA model-composer proprietary platform using MATLAB, Simulink, SystemGenerator, and DSP Builder,
Development of a FPGA-PoC application implementing blockchain consensus algorithms based on cryptonight AES algorithms,
Cloud platform evaluation with RedHat and Mirantis,
Project management with Shape Up and Basecamp.
Heterogeneous computing
Obtaining high compute performance on today’s modern computer architectures requires code that is optimized, power-efficient, and scalable. The demand for high performance continues to increase due to needs in AI, video analytics, data analytics, as well as in traditional high-performance computing (HPC). Modern workload diversity has resulted in a need for architectural diversity; no single architecture is best for every workload. A mix of scalar, vector, matrix, and spatial (SVMS) architectures deployed in CPU, GPU, AI, and FPGA accelerators are required to extract the needed performance.
High-performance computing
Practice of aggregating computing power in a way that delivers much higher performance than typical desktop computers or workstations.
Highly complex computations and data-intensive modeling.
Entire solution stack: compute, data storage systems, networking, management software, and end-user computational applications
High-performance computing as a service (HPCaaS):
Management of HPC resources is a very tedious task for enterprises due to high deployment and maintenance costs.
HPCaaS overcomes this challenge by providing the enterprises processing power on-demand, and
Enables organizations to build their own virtual machines (VMs) delivering advantages such as availability, scalability, agility, flexibility, and reliability.
The content presented here is entirely based on public materials that the IBM Research Europe Zürich and its cloudFPGA team have shared with the scientific community through their profile on Researchgate.
