Performance Monitoring for GPU Accelerated Data Processing

Ref. 2025_007
General-purpose computing on graphics processing units (GPGPU) is increasingly applied in data processing applications. The high degree of parallelism, coupled with memory bandwidth that surpasses that of any CPU, and high-performance interconnects with speeds in the Terabyte/s range make GPUs attractive for large-volume data analytics system.
IBM’s Zurich Research lab offers an internship on the topic of measuring query execution performance on GPU accelerated platforms. The tasks of the internship are to instrument existing data processing code written in C++, performing benchmark measurements, analyzing the results and identify patterns across a range of different queries. The goal is to automate the monitoring part such that benchmarks results can be automatically obtained. The environment is based on NVIDIA’s hardware- and software stack and candidates will be using NVIDIA tools from this environment.
Qualifications

• Current Master’s student (or recent graduate) in Computer Science, Engineering, or a related field.
• Interest in and understanding of systems programming
• Experience in C++
• Familiarity with Linux environments and software development tools (git/GitHub, IDEs, etc.)
• Creativity and outstanding problem-solving ability
  
  

Preferred Qualifications

• Prior experience or knowledge of Database Management Systems (DBMS)
• Experience with Velox and Presto
• Experience using NVIDIA profiling tools
• Excellent oral and written English with good presentation skills.
• Strong interpersonal skills and excellent written and verbal communication.
  
  

Diversity
IBM is committed to diversity at the workplace. With us you will find an open, multicultural environment. Excellent flexible working arrangements enable all genders to strike the desired balance between their professional development and their personal lives.