Pat McGarry Ryft Systems, Inc. Closing Keynote Harnessing the Flood of IoT Data With Heterogeneous Computing at the Edge.

Slides:



Advertisements
Similar presentations
Accessing I/O Devices Processor Memory BUS I/O Device 1 I/O Device 2.
Advertisements

Operating System.
FUTURE TECHNOLOGIES Lecture 13.  In this lecture we will discuss some of the important technologies of the future  Autonomic Computing  Cloud Computing.
GPGPU Introduction Alan Gray EPCC The University of Edinburgh.
1/1/ / faculty of Electrical Engineering eindhoven university of technology Introduction Part 3: Input/output and co-processors dr.ir. A.C. Verschueren.
Basic Input Output System
Big Data Management and Analytics Introduction Spring 2015 Dr. Latifur Khan 1.
© 2009 VMware Inc. All rights reserved Big Data’s Virtualization Journey Andrew Yu Sr. Director, Big Data R&D VMware.
Reconstruction and Analysis on Demand: A Success Story Christopher D. Jones Cornell University, USA.
SANS A Simple Ad hoc Network Simulator Nicolas Burri Roger Wattenhofer Yves Weber Aaron Zollinger.
Unlock Your Data Rich connectivity Robust data integration Enterprise-class manageability Deliver Relevant Information Intuitive design environment.
1© Copyright 2015 EMC Corporation. All rights reserved. SDN INTELLIGENT NETWORKING IMPLICATIONS FOR END-TO-END INTERNETWORKING Simone Mangiante Senior.
1 3 rd SG13 Regional Workshop for Africa on “ITU-T Standardization Challenges for Developing Countries Working for a Connected Africa” (Livingstone, Zambia,
Operating Systems Concepts 1. A Computer Model An operating system has to deal with the fact that a computer is made up of a CPU, random access memory.
Amadeus Travel Intelligence ‘Monetising’ big data sets
GPGPU overview. Graphics Processing Unit (GPU) GPU is the chip in computer video cards, PS3, Xbox, etc – Designed to realize the 3D graphics pipeline.
Module I Overview of Computer Architecture and Organization.
What is Concurrent Programming? Maram Bani Younes.
U.S. Department of the Interior U.S. Geological Survey David V. Hill, Information Dynamics, Contractor to USGS/EROS 12/08/2011 Satellite Image Processing.
Chapter 3.1:Operating Systems Concepts 1. A Computer Model An operating system has to deal with the fact that a computer is made up of a CPU, random access.
Chapter 3 Memory Management: Virtual Memory
NVDA Preetam Jinka Akhil Kolluri Pavan Naik. Background Graphics processing units (GPUs) Chipsets Workstations Personal computers Mobile devices Servers.
© Copyright 2010 Hewlett-Packard Development Company, L.P. 1 MORE THAN A DIGIT(AL) SHIFT Frank Huynen Director Strategy & Innovation.
A Survey of Mobile Cloud Computing Application Models
Operating System. Architecture of Computer System Hardware Operating System (OS) Programming Language (e.g. PASCAL) Application Programs (e.g. WORD, EXCEL)
CPU BASICS, THE BUS, CLOCKS, I/O SUBSYSTEM Philip Chan.
Tufts Wireless Laboratory School Of Engineering Tufts University “Network QoS Management in Cyber-Physical Systems” Nicole Ng 9/16/20151 by Feng Xia, Longhua.
Computer Graphics Graphics Hardware
Frontiers in Massive Data Analysis Chapter 3.  Difficult to include data from multiple sources  Each organization develops a unique way of representing.
GPU Architecture and Programming
Server Virtualization
Big Data Analytics Large-Scale Data Management Big Data Analytics Data Science and Analytics How to manage very large amounts of data and extract value.
Accessing I/O Devices Processor Memory BUS I/O Device 1 I/O Device 2.
Chapter 2 Processes and Threads Introduction 2.2 Processes A Process is the execution of a Program More specifically… – A process is a program.
Neural Networks in Computer Science n CS/PY 231 Lab Presentation # 1 n January 14, 2005 n Mount Union College.
RISC and CISC. What is CISC? CISC is an acronym for Complex Instruction Set Computer and are chips that are easy to program and which make efficient use.
Parallel Computing.
This material exempt per Department of Commerce license exception TSU Xilinx On-Chip Debug.
Simplifying Cloud Connectivity for Your Clients Presenter: Tom SharkeyTom Sharkey December 8,
What we know or see What’s actually there Wikipedia : In information technology, big data is a collection of data sets so large and complex that it.
© 2015 IBM Corporation IBM PureApplication Executive Symposium Diego Segre Vice President, Middleware, Break down the barriers to digital.
CS4315A. Berrached:CMS:UHD1 Introduction to Operating Systems Chapter 1.
GPGPU introduction. Why is GPU in the picture Seeking exa-scale computing platform Minimize power per operation. – Power is directly correlated to the.
Internet of Things. Creating Our Future Together.
WHAT'S THE DIFFERENCE BETWEEN A WEB APPLICATION STREAMING NETWORK AND A CDN? INSTART LOGIC.
TWOJA CYFROWA PRZYSZŁOŚĆ. JUŻ DZISIAJ. Christoph F. Strnadl CTO Central & Eastern Europe 11 May 2016.
Operating Systems p.describe the characteristics of knowledge-based systems; q.describe the purpose of operating systems; r.describe the characteristics.
Advanced Operating Systems CS6025 Spring 2016 Processes and Threads (Chapter 2)
CT101: Computing Systems Introduction to Operating Systems.
Designing Cisco Data Center Unified Fabric
INDUSTRY 4.0: FROM THINGS TO OUTCOMES
Learn how the cloud is accelerating network transformation
Stamatia Bibi1, Dimitris Katsaros2, Panayiotis Bozanis2
Preparing for the Internet of Things 50 Trillion Gigabyte Challenge
Organizations Are Embracing New Opportunities
What is it ? …all via a single, proven Platform-as-a-Service.
Mobile App Development
OPERATING SYSTEMS CS3502 Fall 2017
Enabling machine learning in embedded systems
Grid Computing.
DevOps – Test Automation for IOTs
A Must to Know - Testing IoT
DeFacto Planning on the Powerful Microsoft Azure Platform Puts the Power of Intelligent and Timely Planning at Any Business Manager’s Fingertips Partner.
Chapter 2: System Structures
Operating Systems.
Specialized Cloud Architectures
TrinityIoT Premises Monitoring.
LO2 – Understand Computer Software
Big DATA.
Analytics, BI & Data Integration
Presentation transcript:

Pat McGarry Ryft Systems, Inc. Closing Keynote Harnessing the Flood of IoT Data With Heterogeneous Computing at the Edge

SOURCE: IDC WORLDWIDE IoT TAXONOMY, 2015, WORLDWIDE INTERNET OF THINGS FORECAST, 2015–2020 The Internet of Things (IoT) will trigger the single biggest IT shift since the Internet $1.7+ BILLION BILLION BILLION TRILLION TRILLION $1.7+ BILLION BILLION BILLION TRILLION TRILLION Connected Applications Devices GBs Data Market People Connected Applications Devices GBs Data Market People

 Real-time insights as events occur, close to the source of data  Analysis of data from a range of IoT devices— video, mobile, batch stores, etc.—together  Ultra small & efficient analytics infrastructure  Easy to deploy, use & maintain systems  Low operational costs  No security or performance trade-offs IoT is exacerbating the widening data analytics technology divide. REQUIREMENTS  Persistent compute/IO/storage bottlenecks  Data analyzed in silos  Data movement & ETL delays  Sprawling inefficient analytics infrastructures  Persistent data privacy & security issues REALITY

The reason? Actionable intelligence from IoT is trapped in analysis platforms built on 70-year old architectures.

Heterogeneous Computing is the answer… SOURCES: BLOOMBERG BUSINESS, THE PLATFORM Heterogeneous computing refers to systems that use more than one kind of processor or cores. These systems gain performance or energy efficiency not just by adding the same type of processors, but by adding dissimilar processors, usually incorporating specialized processing capabilities to handle particular tasks.

…because optimal performance & efficiency demands the right “engine” for the job. CPUsFPGA General purpose computing Sequential in nature Non-deterministic performance Interrupts Memory allocation Problems are broken into a sequence of operations and processed serially Not general purpose Purpose built algorithms Can be reprogramed via firmware Best at data-heavy analysis Search, fuzzy search, image and video analysis, deep learning Inherently massively parallel Can execute many hardware-parallel operations in one clock cycle More output with less power Can complete the same problem at 100X the performance of x86 GPUs Some general purpose computing Can excel at certain complex algorithms Best for image rendering, some image analysis Generally more parallel than CPUs, since GPUs have more cores Generally more power efficient than CPU

Performance CPUFPGAGPU Open API CPUFPGAGPU ….but we still need an open, easy-to-use approach with a business-centric, compute-agnostic open API.

Performance A compute-agnostic Open API example in C:

Performance Non-programmatic interfaces are just as simple: $ ryftprim -n 1 -p fhs -f input.txt -od results.txt -q "(RAW_TEXT CONTAINS \"Michelle\")" -w 16 -d 2 -oi index.txt -v

Open APIs must become prevalent, for all analytics.

 Real-time analysis of the range of data types—video, image, & text— whether IoT-generated or traditionally generated, required to shape a problem or experience at the edge was not possible  Behavioral data could not be analyzed along with traditional data  Data movement, ETL & indexing bogged down the network & slowed analysis  Localized analysis consumed valuable space & resources  Real-time analysis of the range of data types—video, image, & text— whether IoT-generated or traditionally generated, required to shape a problem or experience at the edge was not possible  Behavioral data could not be analyzed along with traditional data  Data movement, ETL & indexing bogged down the network & slowed analysis  Localized analysis consumed valuable space & resources ANALYSIS AT THE EDGE Make informed business decisions by analyzing all your data, immediately.

 Instant analysis of on-location video, text, & sensor data together as it happens allows unprecedented personalization  Immediate insight into newly arriving IoT data correlated to legacy data  Data can be analyzed locally to increase responsiveness & reduce network load  No data movement or ETL = no barriers to real-time insights  Small, easy-to-maintain infrastructure brings powerful analytics to the network edge at a fraction of the space, maintenance & energy  Solve business problems using business-centric, open APIs instead of developing complex computing algorithms  Instant analysis of on-location video, text, & sensor data together as it happens allows unprecedented personalization  Immediate insight into newly arriving IoT data correlated to legacy data  Data can be analyzed locally to increase responsiveness & reduce network load  No data movement or ETL = no barriers to real-time insights  Small, easy-to-maintain infrastructure brings powerful analytics to the network edge at a fraction of the space, maintenance & energy  Solve business problems using business-centric, open APIs instead of developing complex computing algorithms Heterogeneous computing and open APIs eliminate edge computing bottlenecks. THE SOLUTION

Enabling the Internet of Things with real-time insights into massive data volume & velocity at the edge. Ultra-high-speed, compact, high-efficiency, & heterogeneous Ryft ONE accelerates the delivery of valuable insights from large, dynamic, & diverse data sets at its source exclusively using open APIs.

Questions? Visit Ryft’s IoT Slam virtual exhibit. Download white papers, analyst reports, & videos. Pat McGarry