Tools & Techniques for a cost optimized Analytics & AI pipeline

on YouTube ↗24m

" summary

Intel claims concrete wins: hardware plus software optimizations deliver 4–5x speedups on computer vision and up to ~200x on CPU ML kernels (scikit‑learn, XGBoost). Key tools shown include OpenVINO for cross‑compiling models to edge/FPGA/accelerators, Converge.IO for MLOps, Granulate agent that tunes kernel/runtime on Linux containers to raise throughput and cut node count (deployed at Mintra and Grofers), plus SGX, crypto acceleration, SigOpt and Intel Neural Compressor.

AI-written summary · about

" chapters (7)

1. 0:00Cost-optimized analytics pipelines
2. 2:11Intel's role in optimization
3. 6:50Compute, memory, I/O bottlenecks
4. 9:42Granulate runtime optimization
5. 16:45Hardware-software co-design tuning
6. 21:47Security for data-driven solutions
7. 23:33Collaboration in India

" related talks

• 13mAccelerating AI algorithmsGary Brown · Artificial Intelligence Conference in London 2018
• 50mAI acceleration at the edgeJonathan Lee, Michael Kleiner, Allan Gagnon, Alex White, Mohan Potheri · AWS re:Invent 2023
• 30mAI: unleashing the next waveMilind Hanchinmani · Anthill Inside 2017
• 50mLakshmi Narasimhan of Intel at Machine Learning Developers Summit 2019Lakshmi Narasimhan · Machine Learning Developers Summit 2019
• 22mLeverage AI Kits for Faster AI Solutions Tailored to Your NeedsBrian Rogers, Rob DeRider, Michelle Megarry, Kiana Yup, Bereket Gaguro · Microsoft Build 2023
• 06mCollaboration and openness drive innovation in artificial intelligenceMartin Hall · Strata + Hadoop World 2016 - New York, New York
• 09mThe AI stack from chip to analyticsMa Ziya · Strata Data Conference 2017 - New York, New York
• 04m4 real-world AI deployments at enterprise scaleGadi Singer · O'Reilly Artificial Intelligence Conference 2019 - New York, New York