Inserenten

e-fellows.net Stellenmarkt Jobs & Praktika suchen

25 km

Informationen zur Anzeige:

Master Thesis in Design and Evaluation of SRAM-based Compute-In-Memory in RISC-V based Subsystem
Renningen
Aktualität: 21.06.2024

Anzeigeninhalt:

21.06.2024, Bosch-Gruppe
Renningen
Master Thesis in Design and Evaluation of SRAM-based Compute-In-Memory in RISC-V based Subsystem
Ihre Aufgaben:
SRAM-based Compute-in-Memory (CIM) presents new paradigm in microcontroller subsystem design. RISC-V allows parametrization to enable pre-processing and AI workloads such as filter banks, directional microphones, or recurrent networks, in combination with acceleration offered by SRAM CIM. In this thesis, the interaction between RISC-V cluster and SRAM-based CIM will be evaluated, including interconnect topologies and CIM partitioning. Fast model will be first developed to accelerate the exploration and evaluation of such system, prior to RTL implementation. During your thesis, you will evaluate SRAM-based CIM hardware accelerators for DSP and AI workloads. You will develop a fast model RISC-V based subsystem. Furthermore, you will also integrate these hardware accelerator models into the FAST model. Additionally, you will analyse the performance metrics of the subsystem and optimise the components. You will also instantiate an SoC template to implement this subsystem and validate its functionality. Last but not least, you will gain experience and collaborate in a cross-functional team spanning algorithm, deployment, and hardware. Contact & Additional information Social counselling and intermediary service for care services Discounts for employees
Das bringen Sie mit:
Master Thesis in Design and Evaluation of SRAM-based Compute-In-Memory in RISC-V based Subsystem Job ID REF230864F LocationRenningen , Germany Fields of workResearch Join asNot Applicable Job typeFull-time Start DateAccording to arrangement Apply now Welcome to Bosch Your tasks Your profile Education: Master studies in the field of Electrical Engineering, Computer Science or comparable Experience and Knowledge: in Digital Systems, Computer Architecture, RTL Design (VHDL/SystemVerilog) Python, as well as in Linux and LaTeX; background knowledge in Neural Networks Personality and Working Practice: independent, reliable and motivated Languages: very good in English Ahmet Erozan (Functional Department) This location offers

Berufsfeld

Bundesland

Standorte

Master Thesis in Design and Evaluation of SRAM-based Compute-In-Memory in RISC-V based Subsystem

Drucken
Teilen
Renningen