Junior Researcher/Post-doc in CPU Workload Analysis and Simulation Methodologies

Junior Researcher/Post-doc in CPU Workload Analysis and Simulation Methodologies

Research scope: performance analysis and simulation methodologies for mobile workloads

Today’s mobile applications feature complex software stacks and user-centric features, such as gesture-based controls, that set them apart from other workload categories. These features not only pose unique challenges at runtime, but also require new benchmarking and simulation methodologies.

The role presents an exciting opportunity to develop novel methodologies for benchmarking, simulating and analyzing contemporary mobile workloads. A successful candidate will have a unique opportunity for cutting-edge research in an unexplored area with high potential for impact both in terms of research (e.g., through publications and presentations) and industry methodologies and products.

Responsibilities

We are looking for motivated researcher or post-doc to join our research activities and contribute in one or more of the following areas:

• Workload profiling, analysis, and bottleneck identification (e.g., using performance counters and/or binary instrumentation techniques) on a real system and on a simulator (e.g., gem5)
• Development of workload simulation methodologies and analysis of their effectiveness
• Modelling and evaluation of new simulator features
• Modelling and evaluation of new microarchitectural features in a simulator
• Modelling and evaluation of heterogeneous architectures (e.g., CPU+GPU/NPU) and heterogeneous cache hierarchies in a simulator
• Working with other researchers and product teams, including taking input and reporting on findings

Requirements

• A PhD in an area related to computer architecture, simulation methodologies, and/or performance analysis and optimization
• Extensive experience working with simulators (e.g., gem5)
• Experiencing bringing up new workloads on a real system and/or in a simulator, tuning and analyzing their performance (e.g., using performance counters and/or binary instrumentation)
• Good understanding of modern CPU architecture and microarchitecture
• Strong communication skills, ability to perform and present detailed analysis of experimental results
• Strong motivation to join a cutting-edge industrial research environment