The Future of Data-Centric Computing

Explore the future of data-centric computing in this 40-minute conference talk by Dr. Bill Harrod from the U.S. Intelligence Advanced Research Projects Activity (IARPA). Dive into cutting-edge research and innovations reshaping data center computing, focusing on addressing challenges in streaming data analytics. Learn about co-design processes, where hardware and software developers collaborate to optimize performance and efficiency. Discover breakthroughs from AMD, Intel, Qualcomm, and university teams tackling real-time data processing and influence maximization. Examine novel approaches like active messages and Fontan chips that could revolutionize data processing. Understand the AGILE Program's goal of providing timely data-analytic results for predicting adversarial events. Gain insights into system-level intelligent mechanisms for handling large, random, time-varying data streams and structures, enabling scalable and efficient execution of dynamic graph analytic applications.

Introduction
Data Analytics Problem
Program Overview
The Problem
SST
Workflows
CoDesign
Summary
Knowledge Graph
Flow Graph
Reference Codes
Target Metric Table
Workflow 4 Influence maximization
Workflow 4 Task graph
Technical approaches
Performers
AMD
Georgia Tech
Indiana University
Challenges