Artur Mariano, BSc, MSc
PhD Candidate
FG Scientific Computing (beim HRZ)
Mornewegstraße 30
D-64293 Darmstadt, Germany
Phone = +351 92 655 35 06
artur.mariano (at)

Short Biography

I am a staff member of the Scientific Computing Group of Technische Universität Darmstadt in Darmstadt, Germany. During 2012, I had a full-time research position at LabCG, a research group in High Performance and Graphics, at the Computer Science Department of the University of Minho, Braga, Portugal. I was also a fortunate FCT - Fundacao para a Ciencia e Tecnologia - grant holder during two years. Prior to joining LabCG, I worked on another research project, that focused on High Performance Image processing. In the meantime I also had the opportunity of doing a research internship at the University of Texas at Austin, Texas, USA. You can check out my CV here.


I run on African coffee, green and black tea and a gallon (about 4L) of water a day. I am passionate about drug-free bodybuilding and nutrition in general. I support organic local farming, local restaurants and micro-businesses as much as possible. I like to learn foreign languages, even though German has changed that perspective a bit. ☺ When it comes to music, I mostly listen to Soul music. I am a very proud fan, supporter and associate member of the biggest Portuguese club, Sporting Clube de Portugal.


My interests revolve around:

• Accelerated (GPU,DSP,FPGA), parallel and low-power high performance computing;
• Algorithms for graph and (point) lattice problems;
• Lattice-based crypto;
• Scheduling, especially in heterogeneous environments;

Current research

My PhD studies are focused on assessing the tractability of algorithms that solve important lattice-based problems, when in the presence of brand new, high-end computing architectures. These algorithms play a major role in lattice-based cryptography, whose security is strictly related with the hardness of the fundamental problems that these algorithms intend to solve. At this point in time, little is known about how efficient implementations can work around the downsides imposed by their exponential asymptotic time and spacial complexities. As a result, the security of the correspondent cryptosystems is not entirely known, especially in terms of how much and what resources are necessary to break them. The main outcome of my studies lies in the knowledge necessary to answer these questions.


• 2014-2015 Thomas Arnreich (BSc at TU Darmstadt)
• 2014-2015 Fábio Correia (MSc at U Minho, intern at ISC, co-advised with Alberto Proença)
• 2014-2015 Cristiano Sousa (MSc at U Minho, co-advised with Alberto Proença)
• 2015-201? Dominique Metz (BSc at TU Darmstadt)
• 2015-201? Hélder Gonçalves (MSc at U Minho, intern at ISC, co-advised with Alberto Proença )

I am currently not actively looking for new students, but if you have a good background in lattices and/or parallel programming, and you're looking forward to work with me, feel free to drop me a message - please attach your CV.