Topics on Compilers Spring Semester Christine Wagner 2011/04/13

Size: px
Start display at page:

Download "Topics on Compilers Spring Semester Christine Wagner 2011/04/13"

Transcription

1 Topics on Compilers Spring Semester 2011 Christine Wagner 2011/04/13

2 Availability of multicore processors Parallelization of sequential programs for performance improvement Manual code parallelization: time-consuming and error-prone process Compiler based automatic parallelization techniques Speculative parallelization 2011/04/13 2

3 But: parallelization difficult due to cross-iteration dependences Different solutions: DOACROSS parallelism techniques Thread level speculation (TLS) Speculative parallelization 2011/04/13 3

4 Speculative parallelization of loops in sequential programs containing dependences Use of: State separation Multiple value prediction Speculative parallelization 2011/04/13 4

5 TLS: Optimistic assumption: no occurring dependence Parallel execution of statements Possible misspeculation worse performance Speculative parallelization 2011/04/13 5

6 Multiple Value Predictions: Distinct versions of following iteration Parallel computation Speculative parallelization 2011/04/13 6

7 One main thread: first iteration Multiple parallel threads: speculative versions of following iteration Executed in separate memory states simultaneously Speculative parallelization 2011/04/13 7

8 Separation of non-speculative and speculative states (main thread parallel threads) Logical division of entire shared memory space: D: non-speculative state; updates only by main thread P: speculative state; communication of results to main thread C: coordinating state; synchronization and tracking Speculative parallelization 2011/04/13 8

9 D and P memories must support stack, global and heap sections since they are used by all threads Each of these sections must provide state separation Stack Separation: Allocation of local variables, accessed through stack pointer Each thread has own stack and stack-pointer Speculative parallelization 2011/04/13 9

10 Global Section Separation: Stored variables only used by main thread Local copies of global variables used by parallel threads Heap Separation: Dynamic memory allocation on one shared heap Logical separation in D, P and C space heaps when heap chunk is allocated by a thread or for coordination Speculative parallelization 2011/04/13 10

11 Speculative parallelization 2011/04/13 11

12 A loop is a good candidate for parallelization if: 1. Loop has frequent loop-carried dependences 2. Values carried by loop dependences are predictable 1. condition: profiling based dependence analysis 2. condition: variable is predictable if its value can be computed through backward data slicing Speculative parallelization 2011/04/13 12

13 Finding parts of a program which are relevant to the value of a variable at a chosen point (slicing criterion) Deleting irrelevant parts Backward slice: statements with some effect on the criterion Forward slice: statements affected by the criterion Speculative parallelization 2011/04/13 13

14 Generate multiple versions: Compute backward data slices of a variable for each different control flow path Create multiple versions of the second iteration Data slice and path information are computed based on profiling trace Speculative parallelization 2011/04/13 14

15 Version confidence: probability of a version being correct VC = path execution frequency prediction confidence Prediction confidence always 1 if data slice is used as prediction code Multiple variables on same path: merge data slices Speculative parallelization 2011/04/13 15

16 But: performance can be limited due to large data slices Computing only partial slice: values of a variable can be computed based on predictions of other variables in original slice Search point where all variables in the slice can either be computed or predicted with high confidence using simple value predictors Speculative parallelization 2011/04/13 16

17 Speculative parallelization 2011/04/13 17

18 If partial slice is used instead of complete data slice: VC = path execution frequency OCBQ OCBQ = overall confidence of boundary queue Speculative parallelization 2011/04/13 18

19 Compiler performs code transformation Speculative parallelization 2011/04/13 19

20 Speculative parallelization 2011/04/13 20

21 Two problems with basic scheme: Small number of versions cover all popular execution paths: VCs of these versions very high Executing versions with small VCs wastes cores Speculative parallelization 2011/04/13 21

22 VC computation relies on path frequency information: Real run: different frequencies due to different inputs which leads to change of VC Some versions less likely to be correct than expected Wastes cores Speculative parallelization 2011/04/13 22

23 Adaptive technique for better use of available cores: Key idea: consider versions with higher VC as candidates for executing additional iterations beyond second iteration Avoid using cores to execute versions with small VCs (threshold) Speculative parallelization 2011/04/13 23

24 Benchmark: ten tested programs Performance analysis: Basic scheme: highest speedups from 1.12x to 1.51x (3 or 4 versions used) Adaptive scheme: highest speedups from 1.18x to 2.33x (7 versions used) Average utilization of cores: 4 Speculative parallelization 2011/04/13 24

25 Overhead analysis: Time overhead: less than 25 % obtained speedups outweigh cost of overhead Speculative parallelization 2011/04/13 25

26 Speculative parallelization technique: Implemented purely in software Using multiple value predictions and State separation Exploits parallelism between consecutive loop iterations Experimental results: average speedup of 1.7x Speculative parallelization 2011/04/13 26

27 Thank you for listening! Please feel free to ask questions! Speculative parallelization 2011/04/13 27

28 Tian, C., Feng, M. and Gupta, R.: Speculative parallelization using state separation and multiple value prediction. Proceedings of ISSM 10, June 5 6, 2010, Toronto, Ontario, Canada. pp Speculative parallelization 2011/04/13 28

Speculative Parallelization Using State Separation and Multiple Value Prediction

Speculative Parallelization Using State Separation and Multiple Value Prediction Speculative Parallelization Using State Separation and Multiple Value Prediction Chen Tian, Min Feng, Rajiv Gupta University of California, CSE Department, Riverside, CA, 92521 {tianc, mfeng, gupta}@cs.ucr.edu

More information

UNIVERSITY OF CALIFORNIA RIVERSIDE. Speculative Parallelization on Multicore Processors

UNIVERSITY OF CALIFORNIA RIVERSIDE. Speculative Parallelization on Multicore Processors UNIVERSITY OF CALIFORNIA RIVERSIDE Speculative Parallelization on Multicore Processors A Dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Computer

More information

Speculative Parallelization of Sequential Loops On Multicores

Speculative Parallelization of Sequential Loops On Multicores Speculative Parallelization of Sequential Loops On Multicores Chen Tian Min Feng Vijay Nagarajan Rajiv Gupta Department of Computer Science and Engineering University of California at Riverside, CA, U.S.A

More information

Supporting Speculative Parallelization in the Presence of Dynamic Data Structures

Supporting Speculative Parallelization in the Presence of Dynamic Data Structures Supporting Speculative Parallelization in the Presence of Dynamic Data Structures Chen Tian, Min Feng, Rajiv Gupta University of California, CSE Department, Riverside, CA, 92521 {tianc, mfeng, gupta}@cs.ucr.edu

More information

A Feasibility Study for Methods of Effective Memoization Optimization

A Feasibility Study for Methods of Effective Memoization Optimization A Feasibility Study for Methods of Effective Memoization Optimization Daniel Mock October 2018 Abstract Traditionally, memoization is a compiler optimization that is applied to regions of code with few

More information

Speculative Parallelization Technology s only constant is CHANGE. Devarshi Ghoshal Sreesudhan

Speculative Parallelization Technology s only constant is CHANGE. Devarshi Ghoshal Sreesudhan Speculative Parallelization Technology s only constant is CHANGE Devarshi Ghoshal Sreesudhan Agenda Moore s law What is speculation? What is parallelization? Amdahl s law Communication between parallely

More information

Shengyue Wang, Xiaoru Dai, Kiran S. Yellajyosula, Antonia Zhai, Pen-Chung Yew Department of Computer Science & Engineering University of Minnesota

Shengyue Wang, Xiaoru Dai, Kiran S. Yellajyosula, Antonia Zhai, Pen-Chung Yew Department of Computer Science & Engineering University of Minnesota Loop Selection for Thread-Level Speculation, Xiaoru Dai, Kiran S. Yellajyosula, Antonia Zhai, Pen-Chung Yew Department of Computer Science & Engineering University of Minnesota Chip Multiprocessors (CMPs)

More information

2 TEST: A Tracer for Extracting Speculative Threads

2 TEST: A Tracer for Extracting Speculative Threads EE392C: Advanced Topics in Computer Architecture Lecture #11 Polymorphic Processors Stanford University Handout Date??? On-line Profiling Techniques Lecture #11: Tuesday, 6 May 2003 Lecturer: Shivnath

More information

Heuristics for Profile-driven Method- level Speculative Parallelization

Heuristics for Profile-driven Method- level Speculative Parallelization Heuristics for Profile-driven Method- level John Whaley and Christos Kozyrakis Stanford University Speculative Multithreading Speculatively parallelize an application Uses speculation to overcome ambiguous

More information

Speculative Parallelization of Sequential Loops on Multicores

Speculative Parallelization of Sequential Loops on Multicores Int J Parallel Prog (2009) 37:508 535 DOI 10.1007/s10766-009-0111-z Speculative Parallelization of Sequential Loops on Multicores Chen Tian Min Feng Vijay Nagarajan Rajiv Gupta Received: 12 January 2009

More information

Static Java Program Features for Intelligent Squash Prediction

Static Java Program Features for Intelligent Squash Prediction Static Java Program Features for Intelligent Squash Prediction Jeremy Singer,, Adam Pocock, Mikel Lujan, Gavin Brown, Nikolas Ioannou, Marcelo Cintra Thread-level Speculation... Aim to use parallel multi-core

More information

Execution-based Prediction Using Speculative Slices

Execution-based Prediction Using Speculative Slices Execution-based Prediction Using Speculative Slices Craig Zilles and Guri Sohi University of Wisconsin - Madison International Symposium on Computer Architecture July, 2001 The Problem Two major barriers

More information

Lecture 9 Dynamic Compilation

Lecture 9 Dynamic Compilation Lecture 9 Dynamic Compilation I. Motivation & Background II. Overview III. Compilation Policy IV. Partial Method Compilation V. Partial Dead Code Elimination VI. Escape Analysis VII. Results Partial Method

More information

Summary: Open Questions:

Summary: Open Questions: Summary: The paper proposes an new parallelization technique, which provides dynamic runtime parallelization of loops from binary single-thread programs with minimal architectural change. The realization

More information

Porting an MPEG-2 Decoder to the Cell Architecture

Porting an MPEG-2 Decoder to the Cell Architecture Porting an MPEG-2 Decoder to the Cell Architecture Troy Brant, Jonathan Clark, Brian Davidson, Nick Merryman Advisor: David Bader College of Computing Georgia Institute of Technology Atlanta, GA 30332-0250

More information

POSH: A TLS Compiler that Exploits Program Structure

POSH: A TLS Compiler that Exploits Program Structure POSH: A TLS Compiler that Exploits Program Structure Wei Liu, James Tuck, Luis Ceze, Wonsun Ahn, Karin Strauss, Jose Renau and Josep Torrellas Department of Computer Science University of Illinois at Urbana-Champaign

More information

Threads SPL/2010 SPL/20 1

Threads SPL/2010 SPL/20 1 Threads 1 Today Processes and Scheduling Threads Abstract Object Models Computation Models Java Support for Threads 2 Process vs. Program processes as the basic unit of execution managed by OS OS as any

More information

CS533 Concepts of Operating Systems. Jonathan Walpole

CS533 Concepts of Operating Systems. Jonathan Walpole CS533 Concepts of Operating Systems Jonathan Walpole Overview Background Criticism of Threads Defense of Threads Compiler Support for Threads Evaluation Conclusion Background: Events Small, static number

More information

Accelerated Library Framework for Hybrid-x86

Accelerated Library Framework for Hybrid-x86 Software Development Kit for Multicore Acceleration Version 3.0 Accelerated Library Framework for Hybrid-x86 Programmer s Guide and API Reference Version 1.0 DRAFT SC33-8406-00 Software Development Kit

More information

Software Speculative Multithreading for Java

Software Speculative Multithreading for Java Software Speculative Multithreading for Java Christopher J.F. Pickett and Clark Verbrugge School of Computer Science, McGill University {cpicke,clump}@sable.mcgill.ca Allan Kielstra IBM Toronto Lab kielstra@ca.ibm.com

More information

Multithreaded Value Prediction

Multithreaded Value Prediction Multithreaded Value Prediction N. Tuck and D.M. Tullesn HPCA-11 2005 CMPE 382/510 Review Presentation Peter Giese 30 November 2005 Outline Motivation Multithreaded & Value Prediction Architectures Single

More information

SEED: A Statically-Greedy and Dynamically-Adaptive Approach for Speculative Loop Execution

SEED: A Statically-Greedy and Dynamically-Adaptive Approach for Speculative Loop Execution 1 SEED: A Statically-Greedy and Dynamically-Adaptive Approach for Speculative Loop Execution Lin Gao, Lian Li, Jingling Xue, Senior Member, IEEE, and Pen-Chung Yew, Fellow, IEEE Abstract Research on compiler

More information

Exploiting Thread-Level Speculative Parallelism with Software Value Prediction

Exploiting Thread-Level Speculative Parallelism with Software Value Prediction Exploiting Thread-Level Speculative Parallelism with Software Value Prediction Abstract Software value prediction (SVP) is an effective and powerful compilation technique helping to expose thread-level

More information

Enabling Loop Parallelization with Decoupled Software Pipelining in LLVM: Final Report

Enabling Loop Parallelization with Decoupled Software Pipelining in LLVM: Final Report Enabling Loop Parallelization with Decoupled Software Pipelining in LLVM: Final Report Ameya Velingker and Dougal J. Sutherland {avelingk, dsutherl}@cs.cmu.edu http://www.cs.cmu.edu/~avelingk/compilers/

More information

CS229 Project: TLS, using Learning to Speculate

CS229 Project: TLS, using Learning to Speculate CS229 Project: TLS, using Learning to Speculate Jiwon Seo Dept. of Electrical Engineering jiwon@stanford.edu Sang Kyun Kim Dept. of Electrical Engineering skkim38@stanford.edu ABSTRACT We apply machine

More information

Virtual Machines. 2 Disco: Running Commodity Operating Systems on Scalable Multiprocessors([1])

Virtual Machines. 2 Disco: Running Commodity Operating Systems on Scalable Multiprocessors([1]) EE392C: Advanced Topics in Computer Architecture Lecture #10 Polymorphic Processors Stanford University Thursday, 8 May 2003 Virtual Machines Lecture #10: Thursday, 1 May 2003 Lecturer: Jayanth Gummaraju,

More information

Parallel Functional Programming Lecture 1. John Hughes

Parallel Functional Programming Lecture 1. John Hughes Parallel Functional Programming Lecture 1 John Hughes Moore s Law (1965) The number of transistors per chip increases by a factor of two every year two years (1975) Number of transistors What shall we

More information

ECE 669 Parallel Computer Architecture

ECE 669 Parallel Computer Architecture ECE 669 Parallel Computer Architecture Lecture 23 Parallel Compilation Parallel Compilation Two approaches to compilation Parallelize a program manually Sequential code converted to parallel code Develop

More information

Exploiting the Behavior of Generational Garbage Collector

Exploiting the Behavior of Generational Garbage Collector Exploiting the Behavior of Generational Garbage Collector I. Introduction Zhe Xu, Jia Zhao Garbage collection is a form of automatic memory management. The garbage collector, attempts to reclaim garbage,

More information

Data Modeling and Databases Ch 10: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich

Data Modeling and Databases Ch 10: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Data Modeling and Databases Ch 10: Query Processing - Algorithms Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Transactions (Locking, Logging) Metadata Mgmt (Schema, Stats) Application

More information

Hyperthreading Technology

Hyperthreading Technology Hyperthreading Technology Aleksandar Milenkovic Electrical and Computer Engineering Department University of Alabama in Huntsville milenka@ece.uah.edu www.ece.uah.edu/~milenka/ Outline What is hyperthreading?

More information

Data Modeling and Databases Ch 9: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich

Data Modeling and Databases Ch 9: Query Processing - Algorithms. Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Data Modeling and Databases Ch 9: Query Processing - Algorithms Gustavo Alonso Systems Group Department of Computer Science ETH Zürich Transactions (Locking, Logging) Metadata Mgmt (Schema, Stats) Application

More information

SPECULATIVE MULTITHREADED ARCHITECTURES

SPECULATIVE MULTITHREADED ARCHITECTURES 2 SPECULATIVE MULTITHREADED ARCHITECTURES In this Chapter, the execution model of the speculative multithreading paradigm is presented. This execution model is based on the identification of pairs of instructions

More information

Abstract Data Structures IB Computer Science. Content developed by Dartford Grammar School Computer Science Department

Abstract Data Structures IB Computer Science. Content developed by Dartford Grammar School Computer Science Department Abstract Data Structures IB Computer Science Content developed by Dartford Grammar School Computer Science Department HL Topics 1-7, D1-4 1: System design 2: Computer Organisation 3: Networks 4: Computational

More information

CSE 544: Principles of Database Systems

CSE 544: Principles of Database Systems CSE 544: Principles of Database Systems Anatomy of a DBMS, Parallel Databases 1 Announcements Lecture on Thursday, May 2nd: Moved to 9am-10:30am, CSE 403 Paper reviews: Anatomy paper was due yesterday;

More information

Compiler Analyses for Improved Return Value Prediction

Compiler Analyses for Improved Return Value Prediction Compiler Analyses for Improved Return Value Prediction Christopher J.F. Pickett Clark Verbrugge {cpicke,clump}@sable.mcgill.ca School of Computer Science, McGill University Montréal, Québec, Canada H3A

More information

Performance Evaluations for Parallel Image Filter on Multi - Core Computer using Java Threads

Performance Evaluations for Parallel Image Filter on Multi - Core Computer using Java Threads Performance Evaluations for Parallel Image Filter on Multi - Core Computer using Java s Devrim Akgün Computer Engineering of Technology Faculty, Duzce University, Duzce,Turkey ABSTRACT Developing multi

More information

Revisiting the Past 25 Years: Lessons for the Future. Guri Sohi University of Wisconsin-Madison

Revisiting the Past 25 Years: Lessons for the Future. Guri Sohi University of Wisconsin-Madison Revisiting the Past 25 Years: Lessons for the Future Guri Sohi University of Wisconsin-Madison Outline VLIW OOO Superscalar Enhancing Superscalar And the future 2 Beyond pipelining to ILP Late 1980s to

More information

Dynamic Memory Scheduling

Dynamic Memory Scheduling Dynamic : Computer Architecture Dynamic Scheduling Overview Lecture Overview The Problem Conventional Solutions In-order Load/Store Scheduling Blind Dependence Scheduling Conservative Dataflow Scheduling

More information

Improving the Practicality of Transactional Memory

Improving the Practicality of Transactional Memory Improving the Practicality of Transactional Memory Woongki Baek Electrical Engineering Stanford University Programming Multiprocessors Multiprocessor systems are now everywhere From embedded to datacenter

More information

UNIVERSITY OF MINNESOTA. Shengyue Wang

UNIVERSITY OF MINNESOTA. Shengyue Wang UNIVERSITY OF MINNESOTA This is to certify that I have examined this copy of a doctoral thesis by Shengyue Wang and have found that it is complete and satisfactory in all respects, and that any and all

More information

Adaptive Speculation in Behavior-Oriented Parallelization

Adaptive Speculation in Behavior-Oriented Parallelization Adaptive Speculation in Behavior-Oriented Parallelization Yunlian Jiang Xipeng Shen Computer Science Department The College of William and Mary, Williamsburg, VA, USA {jiang,xshen}@cs.wm.edu Abstract Behavior-oriented

More information

Agenda Process Concept Process Scheduling Operations on Processes Interprocess Communication 3.2

Agenda Process Concept Process Scheduling Operations on Processes Interprocess Communication 3.2 Lecture 3: Processes Agenda Process Concept Process Scheduling Operations on Processes Interprocess Communication 3.2 Process in General 3.3 Process Concept Process is an active program in execution; process

More information

A STUDY IN THE INTEGRATION OF COMPUTER ALGEBRA SYSTEMS: MEMORY MANAGEMENT IN A MAPLE ALDOR ENVIRONMENT

A STUDY IN THE INTEGRATION OF COMPUTER ALGEBRA SYSTEMS: MEMORY MANAGEMENT IN A MAPLE ALDOR ENVIRONMENT A STUDY IN THE INTEGRATION OF COMPUTER ALGEBRA SYSTEMS: MEMORY MANAGEMENT IN A MAPLE ALDOR ENVIRONMENT STEPHEN M. WATT ONTARIO RESEARCH CENTER FOR COMPUTER ALGEBRA UNIVERSITY OF WESTERN ONTARIO LONDON

More information

Revisiting the Sequential Programming Model for Multi-Core

Revisiting the Sequential Programming Model for Multi-Core Revisiting the Sequential Programming Model for Multi-Core Matthew J. Bridges, Neil Vachharajani, Yun Zhang, Thomas Jablin, & David I. August The Liberty Research Group Princeton University 2 Source: Intel/Wikipedia

More information

Predictive Line Buffer: A fast, Energy Efficient Cache Architecture

Predictive Line Buffer: A fast, Energy Efficient Cache Architecture Predictive Line Buffer: A fast, Energy Efficient Cache Architecture Kashif Ali MoKhtar Aboelaze SupraKash Datta Department of Computer Science and Engineering York University Toronto ON CANADA Abstract

More information

Reconfigurable and Self-optimizing Multicore Architectures. Presented by: Naveen Sundarraj

Reconfigurable and Self-optimizing Multicore Architectures. Presented by: Naveen Sundarraj Reconfigurable and Self-optimizing Multicore Architectures Presented by: Naveen Sundarraj 1 11/9/2012 OUTLINE Introduction Motivation Reconfiguration Performance evaluation Reconfiguration Self-optimization

More information

Why do we care about parallel?

Why do we care about parallel? Threads 11/15/16 CS31 teaches you How a computer runs a program. How the hardware performs computations How the compiler translates your code How the operating system connects hardware and software The

More information

Harp-DAAL for High Performance Big Data Computing

Harp-DAAL for High Performance Big Data Computing Harp-DAAL for High Performance Big Data Computing Large-scale data analytics is revolutionizing many business and scientific domains. Easy-touse scalable parallel techniques are necessary to process big

More information

Dynamic Trace Analysis with Zero-Suppressed BDDs

Dynamic Trace Analysis with Zero-Suppressed BDDs University of Colorado, Boulder CU Scholar Electrical, Computer & Energy Engineering Graduate Theses & Dissertations Electrical, Computer & Energy Engineering Spring 4-1-2011 Dynamic Trace Analysis with

More information

A Caching-Oriented FTL Design for Multi-Chipped Solid-State Disks. Yuan-Hao Chang, Wei-Lun Lu, Po-Chun Huang, Lue-Jane Lee, and Tei-Wei Kuo

A Caching-Oriented FTL Design for Multi-Chipped Solid-State Disks. Yuan-Hao Chang, Wei-Lun Lu, Po-Chun Huang, Lue-Jane Lee, and Tei-Wei Kuo A Caching-Oriented FTL Design for Multi-Chipped Solid-State Disks Yuan-Hao Chang, Wei-Lun Lu, Po-Chun Huang, Lue-Jane Lee, and Tei-Wei Kuo 1 June 4, 2011 2 Outline Introduction System Architecture A Multi-Chipped

More information

Copyright IEEE. Citation for the published paper:

Copyright IEEE. Citation for the published paper: Copyright IEEE. Citation for the published paper: Title: Author: Journal: Year: Vol: Issue: Pagination: URL/DOI to the paper: This material is posted here with permission of the IEEE. Such permission of

More information

1 PROCESSES PROCESS CONCEPT The Process Process State Process Control Block 5

1 PROCESSES PROCESS CONCEPT The Process Process State Process Control Block 5 Process Management A process can be thought of as a program in execution. A process will need certain resources such as CPU time, memory, files, and I/O devices to accomplish its task. These resources

More information

Exploring Speculative Parallelism in SPEC2006

Exploring Speculative Parallelism in SPEC2006 Exploring Speculative Parallelism in SPEC2006 Venkatesan Packirisamy, Antonia Zhai, Wei-Chung Hsu, Pen-Chung Yew and Tin-Fook Ngai University of Minnesota, Minneapolis. Intel Corporation {packve,zhai,hsu,yew}@cs.umn.edu

More information

Compiler Optimization of Memory-Resident Value Communication Between Speculative Threads

Compiler Optimization of Memory-Resident Value Communication Between Speculative Threads Compiler Optimization of Memory-Resident Value Communication Between Speculative Threads Antonia Zhai, Christopher B. Colohan, J. Grery Steffan, and Todd C. Mowry School of Computer Science Carnegie Mellon

More information

Optimising Multicore JVMs. Khaled Alnowaiser

Optimising Multicore JVMs. Khaled Alnowaiser Optimising Multicore JVMs Khaled Alnowaiser Outline JVM structure and overhead analysis Multithreaded JVM services JVM on multicore An observational study Potential JVM optimisations Basic JVM Services

More information

AC59/AT59/AC110/AT110 OPERATING SYSTEMS & SYSTEMS SOFTWARE DEC 2015

AC59/AT59/AC110/AT110 OPERATING SYSTEMS & SYSTEMS SOFTWARE DEC 2015 Q.2 a. Explain the following systems: (9) i. Batch processing systems ii. Time sharing systems iii. Real-time operating systems b. Draw the process state diagram. (3) c. What resources are used when a

More information

ALL the assignments (A1, A2, A3) and Projects (P0, P1, P2) we have done so far.

ALL the assignments (A1, A2, A3) and Projects (P0, P1, P2) we have done so far. Midterm Exam Reviews ALL the assignments (A1, A2, A3) and Projects (P0, P1, P2) we have done so far. Particular attentions on the following: System call, system kernel Thread/process, thread vs process

More information

Kernel Level Speculative DSM

Kernel Level Speculative DSM Motivation Main interest is performance, fault-tolerance, and correctness of distributed systems Present our ideas in the context of a DSM system We are developing tools that Improve performance Address

More information

Continuous Object Access Profiling and Optimizations to Overcome the Memory Wall and Bloat

Continuous Object Access Profiling and Optimizations to Overcome the Memory Wall and Bloat Continuous Object Access Profiling and Optimizations to Overcome the Memory Wall and Bloat Rei Odaira, Toshio Nakatani IBM Research Tokyo ASPLOS 2012 March 5, 2012 Many Wasteful Objects Hurt Performance.

More information

Advance Operating Systems (CS202) Locks Discussion

Advance Operating Systems (CS202) Locks Discussion Advance Operating Systems (CS202) Locks Discussion Threads Locks Spin Locks Array-based Locks MCS Locks Sequential Locks Road Map Threads Global variables and static objects are shared Stored in the static

More information

ronny@mit.edu www.cag.lcs.mit.edu/scale Introduction Architectures are all about exploiting the parallelism inherent to applications Performance Energy The Vector-Thread Architecture is a new approach

More information

Ordered Read Write Locks for Multicores and Accelarators

Ordered Read Write Locks for Multicores and Accelarators Ordered Read Write Locks for Multicores and Accelarators INRIA & ICube Strasbourg, France mariem.saied@inria.fr ORWL, Ordered Read-Write Locks An inter-task synchronization model for data-oriented parallel

More information

Illusionist: Transforming Lightweight Cores into Aggressive Cores on Demand

Illusionist: Transforming Lightweight Cores into Aggressive Cores on Demand Illusionist: Transforming Lightweight Cores into Aggressive Cores on Demand A. Ansari, S. Feng, S. Gupta, J. Torrellas, S. Mahlke HPCA - 2013 University of Illinois University of Michigan June 28, 2013.

More information

NOW that the microprocessor industry has shifted its

NOW that the microprocessor industry has shifted its IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 18, NO. 8, AUGUST 2007 1 CMP Support for Large and Dependent Speculative Threads Christopher B. Colohan, Anastasia Ailamaki, Member, IEEE Computer

More information

The Evaluation of Parallel Compilers and Trapezoidal Self- Scheduling

The Evaluation of Parallel Compilers and Trapezoidal Self- Scheduling The Evaluation of Parallel Compilers and Trapezoidal Self- Scheduling Will Smith and Elizabeth Fehrmann May 23, 2006 Multiple Processor Systems Dr. Muhammad Shaaban Overview Serial Compilers Parallel Compilers

More information

Chapter 3 Instruction-Level Parallelism and its Exploitation (Part 5)

Chapter 3 Instruction-Level Parallelism and its Exploitation (Part 5) Chapter 3 Instruction-Level Parallelism and its Exploitation (Part 5) ILP vs. Parallel Computers Dynamic Scheduling (Section 3.4, 3.5) Dynamic Branch Prediction (Section 3.3, 3.9, and Appendix C) Hardware

More information

Branch Prediction. CS441 System Architecture Jeffrey Waldron Caleb Hellickson

Branch Prediction. CS441 System Architecture Jeffrey Waldron Caleb Hellickson Branch Prediction CS441 System Architecture Jeffrey Waldron Caleb Hellickson What is Branch Prediction? Given a tree, try to predict how many branches it will grow as well as the leave fall pattern What

More information

Uni-Address Threads: Scalable Thread Management for RDMA-based Work Stealing

Uni-Address Threads: Scalable Thread Management for RDMA-based Work Stealing Uni-Address Threads: Scalable Thread Management for RDMA-based Work Stealing Shigeki Akiyama, Kenjiro Taura The University of Tokyo June 17, 2015 HPDC 15 Lightweight Threads Lightweight threads enable

More information

Karma: Know-it-All Replacement for a Multilevel cache

Karma: Know-it-All Replacement for a Multilevel cache Karma: Know-it-All Replacement for a Multilevel cache Gala Yadgar Computer Science Department, Technion Assaf Schuster Computer Science Department, Technion Michael Factor IBM Haifa Research Laboratories

More information

ECE 695 Numerical Simulations Lecture 3: Practical Assessment of Code Performance. Prof. Peter Bermel January 13, 2017

ECE 695 Numerical Simulations Lecture 3: Practical Assessment of Code Performance. Prof. Peter Bermel January 13, 2017 ECE 695 Numerical Simulations Lecture 3: Practical Assessment of Code Performance Prof. Peter Bermel January 13, 2017 Outline Time Scaling Examples General performance strategies Computer architectures

More information

Project 0: Implementing a Hash Table

Project 0: Implementing a Hash Table Project : Implementing a Hash Table CS, Big Data Systems, Spring Goal and Motivation. The goal of Project is to help you refresh basic skills at designing and implementing data structures and algorithms.

More information

Dynamic Performance Tuning for Speculative Threads

Dynamic Performance Tuning for Speculative Threads Dynamic Performance Tuning for Speculative Threads Yangchun Luo, Venkatesan Packirisamy, Nikhil Mungre, Ankit Tarkas, Wei-Chung Hsu, and Antonia Zhai Dept. of Computer Science and Engineering Dept. of

More information

Job Re-Packing for Enhancing the Performance of Gang Scheduling

Job Re-Packing for Enhancing the Performance of Gang Scheduling Job Re-Packing for Enhancing the Performance of Gang Scheduling B. B. Zhou 1, R. P. Brent 2, C. W. Johnson 3, and D. Walsh 3 1 Computer Sciences Laboratory, Australian National University, Canberra, ACT

More information

Dynamic Simultaneous Multithreaded Architecture

Dynamic Simultaneous Multithreaded Architecture Dynamic Simultaneous Multithreaded Architecture Daniel Ortiz-Arroyo and Ben Lee School of Electrical Engineering and Computer Science Oregon State University {dortiz, benl}@ece.orst.edu Abstract This paper

More information

Superscalar Processors

Superscalar Processors Superscalar Processors Superscalar Processor Multiple Independent Instruction Pipelines; each with multiple stages Instruction-Level Parallelism determine dependencies between nearby instructions o input

More information

Loop Selection for Thread-Level Speculation

Loop Selection for Thread-Level Speculation Loop Selection for Thread-Level Speculation Shengyue Wang, Xiaoru Dai, Kiran S Yellajyosula Antonia Zhai, Pen-Chung Yew Department of Computer Science University of Minnesota {shengyue, dai, kiran, zhai,

More information

NDSeq: Runtime Checking for Nondeterministic Sequential Specs of Parallel Correctness

NDSeq: Runtime Checking for Nondeterministic Sequential Specs of Parallel Correctness EECS Electrical Engineering and Computer Sciences P A R A L L E L C O M P U T I N G L A B O R A T O R Y NDSeq: Runtime Checking for Nondeterministic Sequential Specs of Parallel Correctness Jacob Burnim,

More information

Parallel Programming Concepts. Parallel Algorithms. Peter Tröger

Parallel Programming Concepts. Parallel Algorithms. Peter Tröger Parallel Programming Concepts Parallel Algorithms Peter Tröger Sources: Ian Foster. Designing and Building Parallel Programs. Addison-Wesley. 1995. Mattson, Timothy G.; S, Beverly A.; ers,; Massingill,

More information

Chapter 03. Authors: John Hennessy & David Patterson. Copyright 2011, Elsevier Inc. All rights Reserved. 1

Chapter 03. Authors: John Hennessy & David Patterson. Copyright 2011, Elsevier Inc. All rights Reserved. 1 Chapter 03 Authors: John Hennessy & David Patterson Copyright 2011, Elsevier Inc. All rights Reserved. 1 Figure 3.3 Comparison of 2-bit predictors. A noncorrelating predictor for 4096 bits is first, followed

More information

A Does Dynamic and Speculative Parallelization Enable Advanced Parallelizing and Optimizing Code Transformations?

A Does Dynamic and Speculative Parallelization Enable Advanced Parallelizing and Optimizing Code Transformations? A Does Dynamic and Speculative Parallelization Enable Advanced Parallelizing and Optimizing Code Transformations? Philippe Clauss and Alexandra Jimborean, CAMUS group, INRIA, LSIIT, University of Strasbourg

More information

Bumyong Choi Leo Porter Dean Tullsen University of California, San Diego ASPLOS

Bumyong Choi Leo Porter Dean Tullsen University of California, San Diego ASPLOS Bumyong Choi Leo Porter Dean Tullsen University of California, San Diego ASPLOS 2008 1 Multi-core proposals take advantage of additional cores to improve performance, power, and/or reliability Moving towards

More information

ARRAY DATA STRUCTURE

ARRAY DATA STRUCTURE ARRAY DATA STRUCTURE Isha Batra, Divya Raheja Information Technology Dronacharya College Of Engineering, Farukhnagar,Gurgaon Abstract- In computer science, an array data structure or simply an array is

More information

Workloads Programmierung Paralleler und Verteilter Systeme (PPV)

Workloads Programmierung Paralleler und Verteilter Systeme (PPV) Workloads Programmierung Paralleler und Verteilter Systeme (PPV) Sommer 2015 Frank Feinbube, M.Sc., Felix Eberhardt, M.Sc., Prof. Dr. Andreas Polze Workloads 2 Hardware / software execution environment

More information

HELIX-RC: An Architecture-Compiler Co-Design for Automatic Parallelization of Irregular Programs

HELIX-RC: An Architecture-Compiler Co-Design for Automatic Parallelization of Irregular Programs HELIX-RC: An Architecture-Compiler Co-Design for Automatic Parallelization of Irregular Programs Simone Campanoni Kevin Brownell Svilen Kanev Timothy M. Jones + Gu-Yeon Wei David Brooks Harvard University

More information

CHAPTER 4 K-MEANS AND UCAM CLUSTERING ALGORITHM

CHAPTER 4 K-MEANS AND UCAM CLUSTERING ALGORITHM CHAPTER 4 K-MEANS AND UCAM CLUSTERING 4.1 Introduction ALGORITHM Clustering has been used in a number of applications such as engineering, biology, medicine and data mining. The most popular clustering

More information

Designing for Performance. Patrick Happ Raul Feitosa

Designing for Performance. Patrick Happ Raul Feitosa Designing for Performance Patrick Happ Raul Feitosa Objective In this section we examine the most common approach to assessing processor and computer system performance W. Stallings Designing for Performance

More information

Using Thread-Level Speculation to Simplify Manual Parallelization

Using Thread-Level Speculation to Simplify Manual Parallelization Using Thread-Level Speculation to Simplify Manual Parallelization Manohar K. Prabhu Stanford University Computer Systems Laboratory Stanford, California 94305 mkprabhu@stanford.edu Kunle Olukotun Stanford

More information

Proceedings of the 2 nd Java TM Virtual Machine Research and Technology Symposium (JVM '02)

Proceedings of the 2 nd Java TM Virtual Machine Research and Technology Symposium (JVM '02) USENIX Association Proceedings of the 2 nd Java TM Virtual Machine Research and Technology Symposium (JVM '02) San Francisco, California, USA August 1-2, 2002 THE ADVANCED COMPUTING SYSTEMS ASSOCIATION

More information

Efficient Architecture Support for Thread-Level Speculation

Efficient Architecture Support for Thread-Level Speculation Efficient Architecture Support for Thread-Level Speculation A THESIS SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF THE UNIVERSITY OF MINNESOTA BY Venkatesan Packirisamy IN PARTIAL FULFILLMENT OF THE

More information

Operating System. Chapter 4. Threads. Lynn Choi School of Electrical Engineering

Operating System. Chapter 4. Threads. Lynn Choi School of Electrical Engineering Operating System Chapter 4. Threads Lynn Choi School of Electrical Engineering Process Characteristics Resource ownership Includes a virtual address space (process image) Ownership of resources including

More information

Lecture 27. Pros and Cons of Pointers. Basics Design Options Pointer Analysis Algorithms Pointer Analysis Using BDDs Probabilistic Pointer Analysis

Lecture 27. Pros and Cons of Pointers. Basics Design Options Pointer Analysis Algorithms Pointer Analysis Using BDDs Probabilistic Pointer Analysis Pros and Cons of Pointers Lecture 27 Pointer Analysis Basics Design Options Pointer Analysis Algorithms Pointer Analysis Using BDDs Probabilistic Pointer Analysis Many procedural languages have pointers

More information

Lecture 20 Pointer Analysis

Lecture 20 Pointer Analysis Lecture 20 Pointer Analysis Basics Design Options Pointer Analysis Algorithms Pointer Analysis Using BDDs Probabilistic Pointer Analysis (Slide content courtesy of Greg Steffan, U. of Toronto) 15-745:

More information

Trace Compilation. Christian Wimmer September 2009

Trace Compilation. Christian Wimmer  September 2009 Trace Compilation Christian Wimmer cwimmer@uci.edu www.christianwimmer.at September 2009 Department of Computer Science University of California, Irvine Background Institute for System Software Johannes

More information

Pointers II. Class 31

Pointers II. Class 31 Pointers II Class 31 Compile Time all of the variables we have seen so far have been declared at compile time they are written into the program code you can see by looking at the program how many variables

More information

Software and Hardware for Exploiting Speculative Parallelism with a Multiprocessor

Software and Hardware for Exploiting Speculative Parallelism with a Multiprocessor Software and Hardware for Exploiting Speculative Parallelism with a Multiprocessor Jeffrey Oplinger, David Heine, Shih-Wei Liao, Basem A. Nayfeh, Monica S. Lam and Kunle Olukotun Computer Systems Laboratory

More information

Supplementary Material for The Generalized PatchMatch Correspondence Algorithm

Supplementary Material for The Generalized PatchMatch Correspondence Algorithm Supplementary Material for The Generalized PatchMatch Correspondence Algorithm Connelly Barnes 1, Eli Shechtman 2, Dan B Goldman 2, Adam Finkelstein 1 1 Princeton University, 2 Adobe Systems 1 Overview

More information

Chapter 4: Multithreaded Programming

Chapter 4: Multithreaded Programming Chapter 4: Multithreaded Programming Silberschatz, Galvin and Gagne 2013! Chapter 4: Multithreaded Programming Overview Multicore Programming Multithreading Models Threading Issues Operating System Examples

More information

Towards Parallelizing the Layout Engine of Firefox* HotPar 10

Towards Parallelizing the Layout Engine of Firefox* HotPar 10 Towards Parallelizing the Layout Engine of Firefox* HotPar 10 Carmen Badea, UC Irvine Mohammad R. Haghighat, Intel Corporation Alex Nicolau, UC Irvine Alex Veidenbaum, UC Irvine June 14 th, 2010 *This

More information

Runtime. The optimized program is ready to run What sorts of facilities are available at runtime

Runtime. The optimized program is ready to run What sorts of facilities are available at runtime Runtime The optimized program is ready to run What sorts of facilities are available at runtime Compiler Passes Analysis of input program (front-end) character stream Lexical Analysis token stream Syntactic

More information