Intel and the Role of Open Source
summary
Intel describes concrete open source work to make Hadoop and Spark faster and more secure: optimized encryption reduces setup overhead from 50% to 3%, yielding a 16x performance improvement for encrypt/decrypt pipelines on Intel platforms. Spark runs in-memory analytics for near‑real‑time tasks (searching multiple video feeds to find a wandering patient, tailoring digital signage). Ongoing collaborations with Databricks and AMPLab target GraphX, machine‑learning libraries and R optimizations to run efficiently on Intel hardware.
AI-written summary · about
chapters (5)
related talks
14mKeynote: Open Source, Better, Faster, Stronger
24mIntel and Open Source: Bringing AI Everywhere
06mDemocratizing Big Data and Hadoop-Based Platforms and Solutions
19mThe Software Defined World
06mOpen by design, open for data
17mThe Importance of Linux at Intel
13mKeynote: Open Source: The New Normal
32mBig Data With Open Source Solutions