Student (m/f/d): Implementation of hardware Key-Value data structure for O(1) lookup

IMMS Institut für Mikroelektronik- und Mechatronik-Systeme gemeinnützige GmbH (IMMS GmbH); Ilmenau; On-site; EUR 8,928 - 13,392; Full time

Benefits offered by this job

Attraktiver Arbeitsplatz in einem modernen Forschungsinstitut

Direkte Zusammenarbeit mit der Industrie

Flexibles und kreatives Team

Job summary

IMMS GmbH in Ilmenau sucht einen Studenten (m/w/d) für ein Praktikum zur Implementierung einer hardwarebasierten Schlüssel-Wert-Datenstruktur mit O(1) Lookup-Zeiten. In diesem Projekt wird an innovativen Algorithmen wie Cuckoo Hashing und Hopscotch Hashing geforscht und diese in Hardware umgesetzt.

Der Arbeitsplatz ist modern ausgestattet und erlaubt kreatives Arbeiten an herausfordernden Themen, direkt an der Schnittstelle zwischen Universität und Industrie.

Wenn Sie ein kreativer, motivierter Student sind und über Kenntnisse in Python und Verilog verfügen, freuen wir uns auf Ihre Bewerbung.

Qualifications

• Kenntnisse in Python und (System) Verilog erforderlich.
• Grundkenntnisse in Git.
• Grundverständnis der Automatentheorie und Hardware-Beschreibungssprachen von Vorteil.

Responsibilities

• Forschung an potenziellen Algorithmen und/oder Datenstrukturen.
• Implementierung von Hardwaremodulen.
• Dokumentation der entwickelten Hardware.

Skills

Python

System Verilog

Git

Automata theory

Education

Studentenstatus

Job description

Student (m/f/d): Implementation of hardware Key-Value data structure for O(1) lookup

Place of work: Erfurt

Team: Microelectronics

Career level: Mandatory Internship, Thesis or Student Research Assistant

Research field: Integrated sensor systems

Time scope: By agreement

Start: As soon as possible

Application deadline: 2026-12-31

Reference number: IMMS_STUD_ME_0526

Project Overview

Many hardware systems rely on on-chip SRAM for key/value structures, typically using sorted tables and binary search. However, the O(log N) lookup time bottlenecks wide parallel interfaces, and maintaining sorted order requires costly host CPU intervention during insertions. This project aims to investigate and implement fully hardware‑managed alternatives that provide O(1) average lookup times, low deterministic latency, and eliminate CPU‑driven reordering. Since insertions are infrequent, increased complexity during updates is an acceptable trade‑off for faster, predictable lookups.

Candidate algorithms include Cuckoo Hashing, Hopscotch Hashing, and Bloom‑filter‑augmented structures. The chosen algorithms will first be prototyped and evaluated in Python and then implemented as synthesizable hardware using noRTL, a Python‑embedded DSL that automatically converts sequential algorithms into finite state machines and Verilog code. Finally, the resulting hardware block will be verified and benchmarked within a representative test system consisting of an AHB master, DMA, and an SRAM model.

Responsibilities

• Research potential algorithms and/or data structures.
• Implement one or more hardware modules.
• Document the developed hardware.

Qualifications

• Knowledge of Python and (System) Verilog.
• Basic knowledge of Git.
• Basic understanding of automata theory and hardware description languages is an advantage.

Benefits

• An attractive workplace in a modern, very well‑equipped and industry‑oriented research institute.
• Work directly at the interface between university and industry.
• Work in a flexible and creative team and on innovative and challenging topics.

Equal Opportunity Statement

We value and promote equal opportunities and are committed to the goals of equality and diversity. Applications for the listed positions and existing working conditions are welcome regardless of gender and/or any physical disabilities. IMMS ensures gender equality in the workplace and strives to increase the proportion of the under‑represented gender. Women are under‑represented in this field. Applications from women are therefore expressly encouraged. People with severe disabilities will be given preference if they have the same qualifications, skills, and professional performance.

Contact

Contact: Eric Schäfer

IMMS Institut für Mikroelektronik- und Mechatronik-Systeme gemeinnützige GmbH (IMMS GmbH)

Ehrenbergstraße 27

98693 Ilmenau

Germany

Website: www.imms.de/students