Arun Karthik Sengottuvel
A5092183293- Parallel Computing and Optimization Techniques
- Low-power high-performance VLSI design
- Image and Video Quality Assessment
- Stochastic Gradient Optimization Techniques
- Advanced Data Compression Techniques
- Video Coding and Compression Technologies
Meta (United States)
2023
Meta has traditionally relied on using CPU-based servers for running inference workloads, specifically Deep Learning Recommendation Models (DLRM), but the increasing compute and memory requirements of these models have pushed company towards specialized solutions such as GPUs or other hardware accelerators. This paper describes company's effort in constructing its first silicon designed recommendation systems; it accelerator architecture platform design, software stack enabling optimizing...
Videos uploaded to Meta's Family-of-Apps are transcoded into multiple bitstreams of various codec formats, resolutions and quality provide the best video across wide variety devices connection bandwidth constraints. On Facebook alone, there more than 4 billion views per day address processing at this scale, we needed a solution that can deliver possible, with shortest amount encoding time — all while being energy efficient, programmable, scalable. In paper, present, Meta Scalable Video...