Eduardo Cabrera Flores is currently an established researcher at the Barcelona Supercomputing Center. Previously, he was a researcher at the Autonomous University of Guerrero, México; a visiting researcher at the Centre for Environmental Policy, Imperial College London; and a research associate at the Institute of Advanced Research Computing (iARC) in the School of Engineering and Computing Sciences, Durham University, UK.
He obtained his PhD in High Performance Computing from the Autonomous University of Barcelona, Spain, and graduated in Computer Engineering from the School of Engineering, at the National Autonomous University of México. Cabrera has over twenty years of experience in parallel and distributed computing, delivering effective, sustainable solutions to social challenges. In this interview, he explains his role in the QUSTom project and much more.
Why did you choose to be a computer engineer and what motivated you to pursue it?
When I was a child, I wanted to learn and understand almost everything (I guess like any other child), get some answers, and become an explorer, actually an archaeologist. I wanted to learn more about the ancient Egyptian, Inca, Mexica, and Maya civilizations. There were, and still are, three most important motivating factors for choosing my profession as an engineer first, and as a researcher later:
- Curiosity and passion for knowledge and understanding,
- Ability in problem-solving, and
- The possibility of making a significant social impact.
I was driven by a deep curiosity about the world and a desire to understand how things work. The thrill of learning something new was a powerful motivator. I wanted and still want to learn. As the Mexican female philosopher, poet, and writer, Sor Juana Inés de la Cruz, stated: “No estudio para saber más, sino para ignorar menos”, which could be literally translated as: “I do not study to know more, but to ignore less”.
I enjoy tackling complex problems and finding solutions to challenging questions. The intellectual challenge and the opportunity to contribute to solving real-world issues are highly appealing to me.
Last but not least, working on projects that could have a significant impact on society, like the QUSTom project, is highly appealing.
Can you provide an overview of your specific role and responsibilities within the QUSTom project, and how it contributes to its overall objectives?
I was the co-leader of task 3.4, the development of 3D HPC inversion software, which aligns with the very first objective of the project: enabling state-of-the-art HPC FWI kernels that reduce the time to solution by at least a factor of 5 compared to standard modelling algorithms. This is an optimisation task, that is, finding the best solution from a set of feasible solutions. In other words, reducing both execution time and memory consumption to increase the productivity of the application.
This includes the following two tasks:
- To characterise the application, in terms of memory usage and execution time, and identifying the main bottlenecks.
- To convert from Python to C programming language the previously identified most consuming part of the application.
Furthermore, I was solely responsible for porting the optimised translated C version to accelerators, specifically NVIDIA GPUs. The overall result was a reduction of the execution time from more than 1,000 seconds to only 74 seconds, a speedup of almost 15 times. Considering that this simulation has to be executed many, many times, in fact hundreds, this 15x speed-up is a huge achievement.
In your opinion, what sets QUSTom apart from other initiatives focused on improving the diagnosis of breast cancer?
First of all, this research involves a heterogeneous group of experts in many different and complementary fields of science: medicine, physics, mathematics, computational science, and software engineering, amongst others. Finally, the type of technology used: the use of sound waves, which is non-invasive, non-ionizing radiation, with high penetration, high resolution, and low cost, makes this research an exceptional approach in the diagnosis of breast cancer.
What are your hopes and expectations for QUSTom’s long-term impact on breast cancer patients and the healthcare community as a whole?
I have very high hopes and expectations, as this research could help patients to be examined regularly with a harmless, reliable, affordable, timely, and cost-effective breast test. It could provide physicians and radiologists with a new or complementary tool to diagnose breast cancer better, quicker, and less expensively.
Do you have any advice for young researchers who would like to follow in your footsteps?
This is a very tricky question, as there isn’t an ideal or unique pathway. You all have to create your own path, perhaps inspired by someone else, but not following the same one, as none of them guarantees success. As the Spanish poet, Antonio Machado, stated: “Caminante, no hay camino, se hace camino al andar”, which could be literally translated as: “Walker, there is no path, the path is made by walking.”
The most important things are curiosity, imagination, and hard work. As Thomas Alva Edison first and Albert Einstein later stated, it takes 1 to 10% inspiration and 99 to 90% perspiration, which could be understood as…hard work!
