Post Doc Researcher - Optimal hypergraph partitioning and applications

Post Doc Researcher - Optimal hypergraph partitioning and applications
Huawei Technologies, Switzerland

Experience
1 Year
Salary
0 - 0
Job Type
Job Shift
Job Category
Traveling
No
Career Level
Telecommute
No
Qualification
Doctorate
Total Vacancies
1 Job
Posted on
Jul 16, 2021
Last Date
Aug 16, 2021
Location(s)

Job Description

Huawei is a leading global information and communications technology (ICT) solutions provider. Through our constant dedication to customer-centric innovation and strong partnerships, we have established leading end-to-end capabilities and strengths across the carrier networks, enterprise, consumer, and cloud computing fields. Our products and solutions have been deployed in over 170 countries serving more than one third of the world’s population.

With 20+ sites across Europe and 1500 researchers, Huawei’s European Research Institute (ERI) oversees fundamental and applied technology research, academic research cooperation projects, and strategic technical planning across our network of European Ramp;D facilities. Huawei’s ERI includes the new Zurich Research Center (ZRC), located in Zurich, Switzerland. A major element of ZRC is a new research laboratory focused on fundamental research in the area of computing systems, spanning new hardware, new software, and new algorithms.

The research work of the lab will be carried out not only by Huawei’s internal research staff but also by our academic research partners in universities across Europe. The lab will provide an open research environment where academics will be encouraged to visit and work on fundamental long-term research alongside Huawei staff in an environment that, like the best universities and research institutes, is open and conducive to such scientific work.

For this new ZRC Laboratory, we are seeking candidates for two industrial post-doc positions as:

Requirements

Researcher on Optimal Hypergraph Partitioning (6 to 12 months)

Hypergraph partitioning is useful across a great variety of domains, including high performance computing, Deep Learning, graph analytics, and Big Data analysis. In their classical use, hypergraphs model data dependences. Its partitioning helps uncover structure that may be exploited to 1) reduce data movement between compute units when processing the data in parallel, 2) reduce data movement from main memory to any single core by conjoining the partitioning with reordering, 3) reduce computational loads by eliminating spurious computational dependences through again partitioning and reordering, and 4) reduce memory requirements of large-scale computations.

Within this project, successful candidates are interested in pushing the envelope on the state-of-the-art in optimal hypergraph partitioning. While this problem is NP-complete, recent research has shown finding optimal bi-partitionings may be obtained with modest computational resources for hypergraphs of non-trivial size. In our vision, optimal partitioning of smaller problems has value as

  • a tool to evaluate state-of-the-art heuristics for hypergraph partitioning, and as
  • an off-line pre-processing step that allows certain important, small-enough problems to be solved to optimality.

The successful candidates have the unique opportunity to work with hypergraphs that naturally arise from the many challenging applications that Huawei, as a leading technology company, envelops. Additionally, we offer the opportunity not only to evaluate the quality of hypergraph partitionings in terms of standard load balance and cost functions, but also to relate solutions to current and future hardware architectures and systems that Huawei develops. If time permits, successful candidates may furthermore modify current state-of-the-art heuristics using insights obtained from optimal partitionings, or propose completely novel heuristics.

Responsibilities:

  • in collaboration with local researchers and external academic researchers, extend current techniques for optimal hypergraph partitioning;
  • evaluate contemporary heuristics for hypergraph partitioning, compare them to optimality, and identify tendencies across and within specific application domains;
  • produce and present research papers at internationally leading conferences and events.

Requirements:

A PhD degree in Mathematics or Computer Science is required. We seek candidates with research interests in one or more of the following areas, which should additionally be demonstrated by a publication track record:

  • hypergraph partitioning and applications thereof,
  • algorithms for branch-and-bound, network flows, and/or their applications,
  • graph algorithms, hypergraph algorithms, and/or (hyper)graph modeling,
  • solid programming experience, preferably in C++.

Creativity and excellent communication ability in English are key. High potential candidates may be considered for an early-career position.

Benefits

At the Zurich Research Center, the successful candidate becomes part of a multicultural team of leading European researchers with expertise spanning from microarchitectures t

Job Specification

Job Rewards and Benefits

Huawei Technologies

Information Technology and Services - Munich, Germany
© Copyright 2004-2024 Mustakbil.com All Right Reserved.