PhD student positions at Mälardalen University

by Björn Lisper, July 4, 2018

2 PhD positions at Mälardalen University in the area of model-based development of parallel software for heterogeneous hardware platforms.

Instruction for applying:

Deadline: 2018-08-18

Mälardalen University (Sweden) announces two PhD student positions

The need for high-performance computing is increasing at a daunting pace.  Computational heterogeneity is the answer. High-performance computing platforms are increasingly becoming heterogeneous, meaning that they contain a combination of different computational units such as CPUs, GPUs, FPGAs, and AI accelerators.

The overall goal of the recently funded project HERO (Heterogeneous systems: software-hardware integration) is to provide a framework that enables model-based development of optimized parallel software, automatic mapping of software to heterogeneous hardware platforms, and provision of automatic hardware acceleration for the developed software.The project will be carried out in cooperation with 5 Swedish industrial partners (Alten, Arcticus, Enea, Unibap, and Volvo).

HERO is divided in three distinct sub-projects (SP):

SP1: Model-based development and resource analysis of parallel software,

SP2: Static code-level resource analysis, and

SP3: Automatic generation of hardware accelerators.

The first PhD student position will focus on modeling and analysis issues, linking SP1 and SP2. The position is in the intersection of model-based software engineering and formal analysis.  The second position will focus on analysis, design and synthesis issues linking SP2 and SP3. The position is in the intersection of compilation, and software-hardware co-design.

Instruction for applying:

The deadline for applying is 2018-08-18.

Contact persons for position #1 (modeling and analysis of heterogeneous embedded systems):

Federico Ciccozzi, [email protected]
Björn Lisper: [email protected]

Contact persons for position #2 (analysis and synthesis of heterogeneous embedded systems):

Masoud Daneshtalab, [email protected]
Björn Lisper: [email protected]