orchestration: turning material breakthroughs into ... orchestration: turning material breakthroughs

Download Orchestration: Turning material breakthroughs into ... Orchestration: Turning material breakthroughs

Post on 18-Oct-2019

2 views

Category:

Documents

0 download

Embed Size (px)

TRANSCRIPT

  • Orchestration: Turning material breakthroughs into application

    performance

    Jeronimo Castrillon (on behalf of the Orchestration team)

    TU Dresden – Cfaed Chair for Compiler Construction

    jeronimo.castrillon@tu-dresden.de

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 2

    Cfaed: Research program (revisited)

    Goal to explore new technologies for electronic

    information processing which overcome the limits of CMOS technology

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 3

    Goal “to explore new technologies for electronic

    information processing which overcome the limits of CMOS technology”

    Cfaed: Research program (revisited)

    Materials & Functions

    Devices & Circuits

    Information Processing

    C M O S

    (in du st ry .fo cu s)

    A !!S ilic on !N an ow ire s

    B !!C ar bo n

    C !!O rg an ic

    D !!B io m ol ec ul ar As se m bl y

    E !!C he m ic al

    F Orchestration

    G Resilience

    H HAEC:

    Highly Adaptive

    Computing Energy6Efficient

    I Biological Systems

    Material research for post-CMOS technologies

    Systems research to handle heterogeneity

    Biology to inspire solutions

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 4

    Orchestration Path

    Mission Turn materials breakthroughs into application

    performance F Orchestration

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 5

    Orchestration Path

    Mission Turn materials breakthroughs into application

    performance F Orchestration

    Materials)Inspired Paths (Paths A2– E)

    HW/SW%Stack Hardware

    Operating2system

    Skeletons App.2Algorithms

    Database application

    High2 Performance Computing

    Mobile2 Communi) cation

    Underlying algorithms: self-adjusting

    Programming abstractions

    Compilers

    Fo rm al 2A na ly si s

    C oo rd in at io nCode synthesis: variant

    generation “Micro-kernels”

    Heterogeneous “template”

    System analysis: modeling, verification &

    reasoning Cross-layer strategies

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 6

    Orchestration Path: the team

    ! Investigators – Prof. Uwe Aßmann – Prof. Franz Baader – Prof. Christel Baier – Prof. Jeronimo Castrillon – Prof. Gerhard Fettweis – Prof. Christof Fetzer – Prof. Jochen Fröhlich

    – Prof. Hermann Härtig – Prof. Wolfgang Lehner – Prof. Wolfgang E. Nagel – Dr. Marcus Völp – Prof. Axel Voigt

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 7

    Orchestration Path: the team (2)

    ! PhD/Postdocs – Nils Asmußen – Andres Goens – Sebastian Haas – Dirk Habich – Immo Huismann – Tomas Karnagel – Sven Karol – Sascha Klüppelholz – Linda Leuschner

    – Matthias Lieber – Siqi Ling – Steffen Märker – Johannes Mey – Benedikt Nöthen – Michael Raitza – Norman Rink – Annett Ungethüm

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 8

    Outline

    ! Introduction

    ! HW/SW stacks

    ! Orchestration stack

    ! Outlook

    ! Concluding remarks

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 9

    What is a stack

    ! Software components that form a complete platform (where you can run applications on) – Web app: Operating system, web server, database,

    programming language

    ! Provide means to handle complexity

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 10

    HW/SW Stack: Example 1

    Source:% https://source.android.com/source/index.html

    Android%sack

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 11

    HW/SW Stack: Example 2

    Source:%Heterogeneous%System%Architecture http://developer.amd.com/resources/heterogen eous=computing/what=is=heterogeneous= system=architecture=hsa/

    HSA%Stack

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 12

    Stacks and wild heterogeneity

    ! Abstract/hide complexity ! Leave as much visible as needed for efficiency

    ! Emerging techs.: rethink established layers – Languages, compilers, operating systems, …

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 13

    Example: Compiler – Runtime

    Compiler

    Runtime environment

    “protocol” assumptions

    0x0

    0xF..F

    Code

    Data

    Heap

    Stack

    Endurance problem: try not to use same locations that often!

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 14

    Example: Flash FTL

    File%system

    Hard%disk

    File%system

    Flash%drive

    Flash%Translation%Layer%(FTL)

    Mapping

    Wear%leveling

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 15

    Outline

    ! Introduction

    ! HW/SW stacks

    ! Orchestration stack

    ! Outlook

    ! Concluding remarks

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 16

    Recall: orchestration stack

    ! Heterogeneous materials – In Cfaed: Along with orchestration

    (see other DMA presentations) ! Research plan

    – Start with heterogeneous CMOS – Tiled for scalability, integrability – Pilot projects with materials as

    they become available Materials)Inspired Paths (Paths A2– E)

    Hardware Operating2system

    Skeletons App.2Algorithms

    Database application

    High2 Performance Computing

    Mobile2 Communi) cation

    Compilers

    Fo rm al 2A na ly si s

    C oo rd in at io n

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 17

    Applications/Drivers

    ! Mobile communication: typically run on heterogeneous

    ! Data-bases: benefits of heterogeneity (see below) ! HPC: high level languages, abstractions for

    scalability, productivity and portability – Adaptive Finite Element Methods (FEM) – Computational Fluid Dynamics (CFD)

    Hardware Operating% system

    Skeletons App.%Algorithms

    Applications

    Compilers

    Fo rm al %A na ly si s

    C oo rd in at io n

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 18

    Applications/Drivers (2)

    ! Adaptive Finite Element Methods (FEM) – Working with multiple-meshes – Communication across subdomains – Requires repartitioning " load balancing

    – Working on ! Portable templatized library ! First results on heterogeneous platforms

    Hardware Operating% system

    Skeletons App.%Algorithms

    Applications

    Compilers

    Fo rm al %A na ly si s

    C oo rd in at io n

    Dendritic%growth [Witkowski15]

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 19

    Applications/Drivers (3)

    ! Computational Fluid Dynamics (CFD) ! Large applications (1010 unknowns) ! Communication/computation is key

    ! Working on ! Heterogeneous implementation ! More versatile fundamental concepts

    Hardware Operating% system

    Skeletons App.%Algorithms

    Applications

    Compilers

    Fo rm al %A na ly si s

    C oo rd in at io n

    [Huismann15]

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 20

    Hardware

    ! Tomahawk: heterogeneous CMOS

    CM /µK

    Hardware Operating% system

    Skeletons App.%Algorithms

    Applications

    Compilers

    Fo rm al %A na ly si s

    C oo rd in at io n

    ! Tiled (with NoC) for scalability ! Allows to integrate wildly

    heterogeneous components ! CoreManager (CM): HW

    support for fine-grained task dispatching

    [Arnold13]

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 21

    Hardware

    ! DB-processor for sorting algorithms Hardware Operating% system

    Skeletons App.%Algorithms

    Applications

    Compilers

    Fo rm al %A na ly si s

    C oo rd in at io n

    Selectivity:

    [Arnold14]

  • Orchestration: Materials for performance

    © Speaker: J. Castrillon Date: September 21, 2015 Page: 22

    Operating Systems

    ! Mic