Research Scientist, Cloud & AI Computing - DPU/GPU/CPU

ByteDance

Responsibilities

About the Team

The ByteDance DPU team is committed to building the computing infrastructure foundation of ByteDance and Volcano Engine Public Cloud. The team is committed to the architecture, development and cutting-edge research of software and hardware technologies (compute/network/storage) for Cloud & AI computing. The key technologies include virtualization Hypervisor, in-house high-performance userspace network protocol, high-speed interconnection, virtual network switch, high-performance storage, GPU virtualization/scheduling etc.

Responsibilities

  • Responsible for DPU software development, performance optimization and stability enhancement.
  • Responsible for new technology exploration, business value analysis and helping with technical proposals.

Qualifications

Minimum Qualification

  • With a research background and formal research training, a PhD is required
  • Familiar with Linux kernel, proficient in kernel subsystems, such as memory, KVM, scheduler, Cgroup, network, storage, file system and other modules, and have relevant practical experience
  • Familiar with x86 & ARM CPU architecture, familiar with CPU underlying performance tuning, understanding of network and storage protocols
  • Familiar with one of the following fields: database systems, big data, distributed storage, serverless computing, AI/ML inference & training, etc。

Preferred Qualification

  • 1~2 years of experience in software-hardware co-design, with specific experience in developing distributed computing systems, high-speed interconnection, and distributed storage
  • Familiar with userspace network technology DPDK, storage technology SPDK
  • Bonus points for understanding virtualization, CXL memory technologies
  • Bonus points for understanding PCIE/RDMA/smartNIC technologies
Read Full Description
Confirmed a day ago. Posted 30+ days ago.

Discover Similar Jobs

Suggested Articles