Unleashing intelligence and data analytics at scale
summary
Intel presents end-to-end results: Hadoop and Spark on the new Xeon Scalable processors deliver 3x performance for big-data workloads, 2.7x for decision-support, and a 93% uplift for high-band workloads; BigDL — a distributed deep‑learning framework for Spark — shows up to 72% training throughput gain and near‑linear scale with 92% scaling efficiency on 16-node clusters (customers report consistent results on 64 nodes). Adopted by JD.com, cloud providers and AWS integrations claim over 100x inference gains on C5 instances.
AI-written summary · about
related talks
13mCombining algorithms and quality data to build better AI platforms
04m4 real-world AI deployments at enterprise scale
04mYuhao Yang Interview
11mHow Companies are Using Spark, and Where the Edge in Big Data Will Be
05mUnleash the Power of Data at Scale
22mBringing Deep Learning into the Big Data Analytics
41mKeynote: How Open Data Science Opens the World of Innovation
50mAI acceleration at the edge