Sambamba : A Runtime System for Online Adaptive Parallelization - PowerPoint PPT Presentation

Feb 10, 2023 •203 likes •379 views

Sambamba : A Runtime System for Online Adaptive Parallelization Clemens Hammacher Kevin Streit Sebastian Hack Andreas Zeller A well-known challenge Modern Legacy Code Hardware The Problem ... The Problem ... Thread 1 Thread 2 Thread 3

Sambamba : A Runtime System for Online Adaptive Parallelization Clemens Hammacher Kevin Streit Sebastian Hack Andreas Zeller
A well-known challenge Modern Legacy Code Hardware
The Problem ...
The Problem ... Thread 1 Thread 2 Thread 3 Thread 4 Thread 5
The Problem ... Thread 1 Thread 2 Thread 3 Thread 4 Thread 5 ...
The Problem ... Thread 1 Thread 2 Thread 3 Thread 4 Thread 5 Thread 1 Thread 2 ... Thread 3 Thread 4
The Problem ... Thread 1 Thread 2 Thread 3 Thread 4 Thread 5 Thread 1 Thread 2 ... Thread 3 Thread 4 • input dependent • machine dependent
Compile-time Analyses Adaptation Parallelization Execution Profiling • Compute data dependences (using DSA) • Generate program dependence graph • Precompute parallelization candidates
Compile-time Analyses Adaptation Parallelization Execution Profiling • Combine local parallel schedules
Compile-time Analyses Adaptation Parallelization Execution Profiling • JIT -compile code and inject into application • Decide which code variant to execute
Compile-time Analyses Adaptation Parallelization Execution Profiling • Profile execution times • Profile branch probabilities
Compile-time Analyses Adaptation Parallelization Execution Profiling • Analyze gathered profiles • Trigger reassessing parallelization decisions
Compile-time Analyses Adaptation Parallelization Execution Profiling • Analyze gathered profiles • Trigger reassessing parallelization decisions
Compile-time Analyses Adaptation Parallelization Execution Profiling • Analyze gathered profiles • Trigger reassessing parallelization decisions
Speedup on CILK example programs 5X 4X 3X 2X 1X 0X All experiments run on an i7 quad-core with HT.
Clemens Hammacher Kevin Streit Sebastian Hack Compile-time Andreas Zeller Analyses Adaptation Parallelization Sambamba Execution Profiling

Recommend

Engineering Self-Adaptive Software Systems with Runtime Models Seminar on QoS Attributes in

Engineering Self-Adaptive Software Systems with Runtime Models Seminar on QoS Attributes in Service- and Cloud-based Systems May 20-25, 2012 Thomas Vogel System Analysis and Modeling Group Hasso Plattner Institute University of Potsdam,

457 views • 43 slides

Adaptation and Abstract Runtime Models 5th Workshop on Software Engineering for Adaptive and

Adaptation and Abstract Runtime Models 5th Workshop on Software Engineering for Adaptive and Self-Managing Systems (SEAMS 2010) Cape Town, South Africa, 3-4 May 2010 Thomas Vogel and Holger Giese System Analysis and Modeling Group Hasso

377 views • 21 slides

Models at Runtime for Adaptive and Self-managing Software Dagstuhl Seminar on Models@run.time

Models at Runtime for Adaptive and Self-managing Software Dagstuhl Seminar on Models@run.time (2011) Holger Giese and Thomas Vogel System Analysis & Modeling Group Hasso Plattner Institute for Software Systems Engineering University of

399 views • 11 slides

A Runtime Environment for Online Processing of Operating System Kernel Events Michael Schbel,

A Runtime Environment for Online Processing of Operating System Kernel Events Michael Schbel, Andreas Polze 7th Intl. Workshop on Dynamic Analysis (WODA) 20. July 2009 Chicago, IL OS Kernel Event Tracing 2 Dynamic Analysis Usage

889 views • 20 slides

A Language for Feedback Loops in Self-Adaptive Systems: Executable Runtime Megamodels 7th Intl.

A Language for Feedback Loops in Self-Adaptive Systems: Executable Runtime Megamodels 7th Intl. Symposium on Software Engineering for Adaptive and Self-Managing Systems (SEAMS 2012) Zurich, Switzerland, June 4-5, 2012 Thomas Vogel and Holger

638 views • 32 slides

Charm++ Migratable Objects + Asynchronous Methods + Adaptive Runtime = Performance +

Charm++ Migratable Objects + Asynchronous Methods + Adaptive Runtime = Performance + Productivity Laxmikant V. Kale Anshu Arya Nikhil Jain Akhil Langer Jonathan Lifflander Harshitha Menon Xiang Ni Yanhua Sun Ehsan Totoni Ramprasad

727 views • 47 slides

Parallel Runtime Environments with Cloud Database: Performance Study for HMM with Adaptive

Parallel Runtime Environments with Cloud Database: Performance Study for HMM with Adaptive Sampling D. Roehm, R. S. Pavel, T. C. Germann, A. L. McPherson and C. Junghans Los Alamos National Laboratory NM, USA May 8, 2015

581 views • 22 slides

Pony: Co-Designing a Type System and a Runtime Sylvan Clebsch March 8th, 2017 Presenting at

Pony: Co-Designing a Type System and a Runtime Sylvan Clebsch March 8th, 2017 Presenting at QCon London My background 25 years of industry programming Electronic trading, online video games, crypto tools, physics modelling, rendering

990 views • 60 slides

Runtime System COMP 524: Programming Languages Based in part on slides and notes by J. Erickson,

Runtime System COMP 524: Programming Languages Based in part on slides and notes by J. Erickson, S. Krishnan, B. Brandenburg, S. Olivier, A. Block, and others What is the Runtime System (RTS)? Language runtime environment. OS view : RTS

962 views • 69 slides

RUNTIME SUPPORT FOR ADAPTIVE SPATIAL PARTITIONING AND INTER-KERNEL COMMUNICATION ON GPUS Yash

RUNTIME SUPPORT FOR ADAPTIVE SPATIAL PARTITIONING AND INTER-KERNEL COMMUNICATION ON GPUS Yash Ukidave , Perhaad Mistry , Charu Kalra , Dana Schaa and David Kaeli Department of Electrical and Computer Engineering Northeastern

502 views • 23 slides

Towards software architecture runtime models for continuous adaptive monitoring Thomas Brand,

13th International Workshop on Models@run.time Towards software architecture runtime models for continuous adaptive monitoring Thomas Brand, Holger Giese 14.10.2018 Agenda Show why it is relevant to investigate and support: Continuous

421 views • 29 slides

OPTIMUM OPTIMUM ADAPTIVE ALGORITHMS ADAPTIVE ALGORITHMS for for SYSTEM IDENTIFICATION SYSTEM

OPTIMUM OPTIMUM ADAPTIVE ALGORITHMS ADAPTIVE ALGORITHMS for for SYSTEM IDENTIFICATION SYSTEM IDENTIFICATION George V. Moustakides Dept. of Computer Engineering & Informatics University of Patras, GREECE Definition of the problem w n x

435 views • 24 slides

Runtime Models as Interfaces for Adapting Software Systems Seminar on Software Engineering for

Runtime Models as Interfaces for Adapting Software Systems Seminar on Software Engineering for Self-Adaptive Systems Schloss Dagstuhl, October 24-29, 2010 Thomas Vogel System Analysis and Modeling Group Hasso Plattner Institute University of

306 views • 14 slides

A Generic Adaptive Runtime Autotuning Framework Isaac Dooley 7th Annual Workshop on Charm++ and

A Generic Adaptive Runtime Autotuning Framework Isaac Dooley 7th Annual Workshop on Charm++ and its Applications Thursday, April 16th, 2009 1 Existing Parallel Programming Models MPI Model Charm++ Model One Thread Per Processor

590 views • 11 slides

The Use of EduPRO System for Adaptive Learning Process Organizing Benefits of adaptive e

Vasyl Stefanyk Precarpathian National University Centre of Distance Learning The Use of EduPRO System for Adaptive Learning Process Organizing Benefits of adaptive e learning system Formation of the individual structure of the material;

396 views • 10 slides

LIRA: Adaptive Contention-Aware Thread Placement for Parallel Runtime Systems Alexander Collins*,

LIRA: Adaptive Contention-Aware Thread Placement for Parallel Runtime Systems Alexander Collins*, Tim Harris , Murray Cole*, Christian Fensch * University of Edinburgh Oracle Labs, UK Heriot Watt University 1 The Problem

925 views • 40 slides

Running PEPPHER benchmarks on top of the StarPU runtime system Cdric Augonnet Nicolas Collin

1 Running PEPPHER benchmarks on top of the StarPU runtime system Cdric Augonnet Nicolas Collin Nathalie Furmento Raymond Namyst Samuel Thibault INRIA Bordeaux, LaBRI, Universit de Bordeaux 22 th January 2011 2 The StarPU runtime system

469 views • 23 slides

Online Phase-Adaptive Data Layout Selection Chengliang Zhang Martin Hirzel Microsoft IBM

Online Phase-Adaptive Data Layout Selection Chengliang Zhang Martin Hirzel Microsoft IBM (former IBM intern) ECOOP, 10 July 2008 1 Problem Statement No training run. Online Phase-Adaptive Data Layout Selection Cache line or page A

344 views • 17 slides

WoT Runtime, Scripting, Bindings Zoltan Kis, Intel WoT Runtime WoT RT Script 1 Things Things

WoT Runtime, Scripting, Bindings Zoltan Kis, Intel WoT Runtime WoT RT Script 1 Things Things Process view WoT RT: one process Bindings: separate processes TDs System APIs / OS Kernel Script n System APIs Socket

359 views • 4 slides

Adaptive PID Controller Yiming Zhao 1 Contents Control system PID controller

Adaptive PID Controller Yiming Zhao 1 Contents Control system PID controller Adaptive PID 2 Control System Open-loop system OUT IN PLANT 3 Control System Closed-loop system/feedback control system input reference

843 views • 20 slides

PROGRAM YOUR OWN RV SYSTEM an exercise in DSL design Klaus Havelund Jet Propulsion Laboratory,

PROGRAM YOUR OWN RV SYSTEM an exercise in DSL design Klaus Havelund Jet Propulsion Laboratory, USA Summer School on Cyber-Physical Systems July 7-10, 2014 Definition of Runtime Verification Definition (Runtime Verification) Runtime

1.23k views • 109 slides

Solid Type System vs Runtime Checks and Unit Tests Vladimir Pavkin Plan Fail Fast concept

Solid Type System vs Runtime Checks and Unit Tests Vladimir Pavkin Plan Fail Fast concept Type Safe Patterns Fail Fast Immediate and visible failure Where can it fail? Handled runtime exceptions & assertions Unhandled runtime failure

1.2k views • 62 slides

Mason Public Schools NWEA Parent Meeting What is NWEA Map Growth? An online computer adaptive

Mason Public Schools NWEA Parent Meeting What is NWEA Map Growth? An online computer adaptive assessment. Computer adaptive means that each student receives different questions depending on how they respond to the previous

809 views • 13 slides

On adaptive regret bounds for non- stochastic bandits Gergely Neu INRIA Lille, SequeL team

On adaptive regret bounds for non- stochastic bandits Gergely Neu INRIA Lille, SequeL team Universitat Pompeu Fabra, Barcelona Online learning and bandits Adaptive bounds in online learning Adaptive bounds for bandits Outline

1.22k views • 75 slides