Overview of Intel Parallel Studio XE

Transcription

1 Overview of Intel Parallel Studio XE Stephen Blair-Chappell 1

2 30-second pitch Intel Parallel Studio XE 2011 Advanced Application Performance What Is It? Suite of tools to develop high performing, robust applications What s Included? Benefit Compilers/Libraries Correctness Analyzer Performance Profiler High performance C/C++/Fortran compilers for x86 CPU s Single tool to detect threading, memory and security errors Easier to use GUI detects performance bottlenecks, code tuning for higher efficiency Maximize application performance and code quality on Windows & Linux 2

3 Three Product Lines for Diverse Needs Essential Performance Advanced Performance Distributed Performance C/C++ developers Microsoft Visual Studio* Take advantage of multicore C++ and Fortran developers Windows* and Linux* High performance, cross platform apps C++ and Fortran developers on Windows* and Linux* High performance MPI clusters intel.com/software/products 3

4 Feature comparison on Windows* Intel Parallel Studio Intel Parallel Studio XE Intel Cluster Studio Parallel Studio XE Composer Static Security Analysis C++ Optimizing Compiler Fortran Optimizing Compiler Advanced compiler optimizations Co-Array Fortran 1 Profile Guided Parallel Debugger Extension Threaded Performance Libraries(Intel IPP) Threading Libraries(Intel TBB, Intel ABB) Threaded Math Libraries (Intel MKL) Intel Parallel Building Blocks Inspector Memory & Thread Analysis Advanced Memory, Thread Analysis Amplifier Hotspot, Concurrency, Locks & Waits Profiling Timeline, Frame Analysis Event based sampling Source view Assembly view Advisor NA NA Threading advice for serial applications Cluster Optimized MPI Library MPI Timeline View, MPI Event View Find MPI deadlocks, data corruption, parameter errors User Interface & Support Visual Studio integration Stand alone graphical interface Command line Basic Advanced Advanced Support: Forums or Premier (includes forums) Premier Premier Premier 1 Node-parallel support only. 4

5 Intel Parallel Studio XE 2011 Powerful tools to create fast, reliable and secure code Phase Productivity Tool Feature Benefit Advance d Build & Debug Intel Composer XE C/C++ and Fortran compilers, performance libraries, and parallel models Application performance, scalability and quality for current multicore and future many-core systems. Advance d Verify Intel Inspector XE Memory & threading error checking tool for higher code reliability & quality Increases productivity and lowers cost, by catching memory and threading defects early Advance d Tune Intel VTune Amplifier XE Performance Profiler to optimize performance and scalability Removes guesswork, saves time, makes it easier to find performance and scalability bottlenecks Combines ease of use with deeper insights. Boost Performance. Code Reliably. Scale Forward. 5

6 Optimizing Compiler and Libraries Intel Composer XE Built-in optimizations and performance libraries make applications go fast Expanded and Simplified Parallelism Options Intel Parallel Studio XE 2011 is a great software development tool for performance-oriented Windows*- based C++ software developers. I achieved an astonishing boost in performance by using Intel Cilk and Array Notation features in my code. If you need performance, try Intel Parallel Studio XE 2011 World Class Fortran Compilers Jorge Martinis Research & Development Engineer BR&E Inc. 6

7 Intel Composer XE 2011 Industry leading multicore C++ and Fortran Compilers & Libraries Performance leadership Most comprehensive multicore and standards support More parallelism-development features, with Intel Parallel Building Blocks Most Fortran 2008 (including Coarrays) and Fortran 2003 features among x86 compilers Advanced vectorization support with technologies like GAP and SIMD pragmas Improved integration of Intel performance libraries: Intel Math Kernel Library, Intel Integrated Performance Primitives s for Sandy Bridge Continued support for earlier processors Compilers and libraries support Intel and compatible processors in a single binary 32-bit and 64-bit support, IA-32 and Intel 64. Windows*, Linux* and Mac* OS X Intel Composer XE 2011 Products Deliver Leadership Performance, Demonstrable with Industry Benchmarks 7

8 Intel Compiler Architecture C++ Front End FORTRAN 2000 Front End Profiler Interprocedural analysis and optimizations: inlining, constant prop, whole program detect, mod/ref, points-to Disambiguation: types, array, pointer, structure, directives Loop optimizations: data deps, prefetch, vectorizer, unroll/interchange/fusion/dist, auto-parallel/openmp Global scalar optimizations: partial redundancy elim, dead store elim, strength reduction, dead code elim Code generation: vectorization, software pipelining, global scheduling, register allocation, code generation 1/18/2011 8

9 Intel Parallel Building Blocks Enables portability, reliability, scalability, simplicity Data parallel and general purpose parallelism solutions Language extensions and library solutions Optimized high-level algorithms and low-level constructs to build custom algorithms Mix and match new parallel models within an application to suit the developer s environment / application and algorithms 9

10 Mixing and Matching Parallel Constructs Low level construct ts high level algorithms MKL IPP Array notations, map functions available Cilk Plus and ArBB Parallel for loops in Cilk, TBB Open SIMD, for loops in OpenMP Intrinsics Data centric Parallelism Higher level of abstraction More control over execution TBB Containers TBB algorithms Cilk spawn OpenMP Parallel invocations in TBB Locks, mutexes, TLS abstractions in TBB General Parallelism Control execution det tails express intent 10

11 Serial and Parallel Code Profiler Intel VTune Amplifier XE Find bottlenecks that limit performance Powerful capabilities, Intuitive setup Without VTune trying to optimize ANY application is simply guessing and wastes valuable engineering time." Tune while viewing source code Steven Peeler Technical Director Ritual Entertainment 11

12 Intel VTune TM Amplifier XE 2011 World s Leading Performance Profiler for IA-32, Intel 64 architectures Next Generation for ease of use, more responsive, yet detailed Quickly locate key bottlenecks Timeline makes it easier to debug thread waiting and balance thread loading Pre-defined advanced analysis makes setup easier Compare results feature to quickly see what changed Shows profiling results in source Command line capabilities for automatic regression tests Ongoing commitment to embrace platform innovation Support for Sandy Bridge Continued support for Intel processors and compatibles 32-bit and 64-bit support Windows* and Linux* Intel VTune Amplifier XE 2011 Delivers Leadership Performance Tuning capabilities to performance developers 12

13 Intel VTune Amplifier XE Tune Applications for Scalable Multicore Performance Fast, Accurate Performance Profiles Hotspot (Statistical call tree) Hardware-Event Based Sampling Thread Profiling Visualize thread interactions on timeline Balance workloads Easy set-up Pre-defined performance profiles Use a normal production build Compatible Microsoft, GCC, Intel compilers C/C++, Fortran, Assembly,.NET Latest Intel processors and compatible processors 1 Find Answers Fast Filter extraneous data View results on the source / assembly Event multiplexing Windows or Linux Visual Studio Integration (Windows) Standalone user i/f and command line 32 and 64-bit 1 IA32 and Intel 64 architectures. Many features work with compatible processors. Event based sampling requires a genuine Intel Processor. 13

14 Amplifier XE 14

15 Intel VTune Amplifier XE Performance Profiler What s New? Preset performance experiments Statistical call tree Concurrency and Locks & Waits Analysis Threading timeline Cross-select and filter results Attach to a running process (Windows only) Better source/asm view Frame based analysis Event multiplexing Linux: No EntireX, root optional for install So What? Easy to get started Faster analysis startup and run Find sync objects limiting scalability Visualize thread transitions Filter out the noise, focus on answers Profile your app without restarting it Easier navigation Isolate data for slow frames More data with fewer runs Fewer conflicts, root only required for EBS install. 15

16 Correctness Analyzer Intel Inspector XE Find memory errors that cause lockups and crashes Find data races and deadlocks that cause lockups and incorrect results It was an easy and fast ramp to start using the Intel Inspector XE 2011 tool, we were able to set the analysis level, obtain a visual interpretation of the collected data, and get helpful information on hidden data races in the code quickly." Find Security Vulnerabilities Alex Migdalski- CEO, CTO of OTRADA Inc 16

17 Intel Inspector XE 2011 Correctness Analyzer. Advancing Application Reliability, Code Quality and Security Enhances productivity, reduces maintenance cost, lowers TCO Powerful dynamic analysis for Memory and Thread error detection Detects memory leaks and memory corruption Pinpoints threading data races and deadlocks Productivity tools Dynamic Instrumentation and time line visualization Detailed error data analysis Support for Sandy Bridge processors Continued support for Intel processors and compatibles 32-bit and 64-bit support Windows* and Linux* Intel Inspector XE 2011 Advancing Application reliability and Security 17

18 Static Security Analysis Easier, faster setup, and ramp to get static analysis results Simple approach to configure and run static analysis Discover and fix defects at any phase of the development cycle Finds over 250 security errors, such as: - Buffer overruns and uninitialized variables - Unsafe library usage and arithmetic overflow - Unchecked input and heap corruption Tracks state associated with issues, even as source evolves and line numbers change Displays problem sets and location of source Provides filters, assignment of priority, and maintenance of problem set state Intuitive standalone GUI & Command Line interface for Windows and Linux Enhanced Application Security Catch defects early in lifecycle C, C++, C#/.NET Fortran for Windows and Linux support, Microsoft Visual Studio* 2005, 2008,

19 Recommended Analysis Sequence Run Static Security Analysis Fix Problems Run Memory Analysis Fix Problems Run Threading Analysis Fix Problems Cleaner code! Use both Static and Dynamic analyses for cleaner code! 1/18/

20 Parallel Advisor Note: Not available in XE version of Parallel Studio 20

21 Intel Parallel Advisor Parallel Advisor analyzes the executing serial program as developers progress through the methodology Survey Target Helps developers focus on the hot call trees and loops as locations to experiment with parallelism. Annotate Sources - Developers insert Parallel Advisor annotations into their sources to describe parallel experiments. Check Suitability Helps developers evaluate the performance of their parallel experiment by displaying the performance projection for each parallel site and how each site s performance impacts the entire program. The Advisor Workflow window guides the developer through the methodology Check Correctness Helps developers by identifying data issues (races) in the parallel experiment. 21

22 Intel Parallel Advisor Survey Target Focuses developers attention to the hot call trees and loops as locations to experiment with parallelism Developers can drill into the source code Top down list of call sites, lines, & loops 22

23 Intel Parallel Advisor Annotate Sources Developers insert annotations into their sources to describe parallel experiments for the hot calls and loops identified by Survey. Assistance with the creation and insertion of Advisor annotations. Parallel Advisor annotations are not a parallel framework. Parallel Advisor annotations are macros. They do not impact the developer s application when Parallel Advisor is not present. required for Suitability and Correctness tools. compatible with all popular fork/join parallel frameworks. 23

24 Intel Parallel Advisor Check Suitability Program impact Scalability of selected site Developers evaluate their parallel experiments based on projected performance improvements Developer s parallel sites How to maximize the performance benefit of the selected site Developer s tasks and locks that were executed within the selected site Help when the developer needs it 24

25 Intel Parallel Advisor Check Correctness Developers evaluate the data sharing problems that prevent correct parallel execution Observation details Problems and Messages Observation problem sets 25

26 Intel Parallel Advisor Summary Report Developers can quickly identify the parallel experiments to focus on based on projected performance gains and the number of correctness problems. Summary of Correctness findings Summary of Suitability performance projection 26

27 Different Approaches to using Parallel Studio Looking for as much help as possible Advisor (not in IPS XE) Prefer to control development steps Four step Methodology I hate graphical user interfaces Command line tools Compiler Amplifier Inspector 27

28 The Four Step Model Analyze Find where to start parallelizing Implement Debug Tune Introduce threads, compile, and debug with Intel Parallel Composer Find threading and memory errors with Intel Parallel Inspector Tune for optimal concurrency usage with Intel Parallel Amplifier 1/18/

29 Workflow with Compiler as a Tool Application Source C/C++/Fortran Compiler Application Binary + Opt Reports Performance Tools Identify hotspots, problems Application Source + Hotspots Compiler in advicemode Advice messages Guided Auto Parallelization Modified Application Source Compiler (extra options) Improved Application Binary Simplifies programmer effort in application tuning 29 1/18/

30 1/18/

31 Intel compilers, associated libraries and associated development tools may include or utilize options that optimize for instruction sets that are available in both Intel and non-intel microprocessors (for example SIMD instruction sets), but do not optimize equally for non-intel microprocessors. In addition, certain compiler options for Intel compilers, including some that are not specific to Intel micro-architecture, are reserved for Intel microprocessors. For a detailed description of Intel compiler options, including the instruction sets and specific microprocessors they implicate, please refer to the Intel Compiler User and Reference Guides under Compiler Options." Many library routines that are part of Intel compiler products are more highly optimized for Intel microprocessors than for other microprocessors. While the compilers and libraries in Intel compiler products offer optimizations for both Intel and Intelcompatible microprocessors, depending on the options you select, your code and other factors, you likely will get extra performance on Intel microprocessors. Intel compilers, associated libraries and associated development tools may or may not optimize to the same degree for non-intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include Intel Streaming SIMD Extensions 2 (Intel SSE2), Intel Streaming SIMD Extensions 3 (Intel SSE3), and Supplemental Streaming SIMD Extensions 3 (Intel SSSE3) instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. While Intel believes our compilers and libraries are excellent choices to assist in obtaining the best performance on Intel and non-intel microprocessors, Intel recommends that you evaluate other compilers and libraries to determine which best meet your requirements. We hope to win your business by striving to offer the best performance of any compiler or library; please let us know if you find we do not. revision # /18/

32 Legal Disclaimer INFORMATION IN THIS DOCUMENT IS PROVIDED AS IS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. INTEL ASSUMES NO LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO THIS INFORMATION INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, reference BunnyPeople, Celeron, Celeron Inside, Centrino, Centrino Atom, Centrino Atom Inside, Centrino Inside, Centrino logo, Cilk, Core Inside, FlashFile, i960, InstantIP, Intel, the Intel logo, Intel386, Intel486, IntelDX2, IntelDX4, IntelSX2, Intel Atom, Intel Atom Inside, Intel Core, Intel Inside, Intel Inside logo, Intel. Leap ahead., Intel. Leap ahead. logo, Intel NetBurst, Intel NetMerge, Intel NetStructure, Intel SingleDriver, Intel SpeedStep, Intel StrataFlash, Intel Viiv, Intel vpro, Intel XScale, Itanium, Itanium Inside, MCS, MMX, Oplus, OverDrive, PDCharm, Pentium, Pentium Inside, skoool, Sound Mark, The Journey Inside, Viiv Inside, vpro Inside, VTune, Xeon, and Xeon Inside are trademarks of Intel Corporation in the U.S. and other countries. *Other names and brands may be claimed as the property of others. Copyright Intel Corporation. 32