The Next Leap in JavaScript Performance

Transcription

1 The Next Leap in JavaScript Performance Mohammad Reza Haghighat Senior Principal Engineer, Intel Corporation October 20, 2014

2 Agenda HTML5 - The New Lingua Franca? Exposing the full power of modern hardware to JavaScript * Bringing Perceptual Computing to the web platform Supporting JavaScript programming in Internet of Things (IoT) Summary 2

3 HTML5 The New Lingua Franca? APPS.exe WEB HTML, Flash * APPS ios *, Android *, Windows * WEB HTML5 Native code PC spiral Web Write once, run on any browser App Stores Walled Gardens Write Once, Run Everywhere New open standards created in the mobile era, such as HTML5, will win on mobile devices. Steve Jobs If you want to do something that is universal, no question, world is going HTML5. Steve Ballmer It looks to me like HTML5 will eventually become a way almost all applications are built, including those on new phones. Eric Schmidt 3

4 Web: The Ubiquitous Software Platform Rich Capabilities & Content Big Data and the Application Model of the Future Social Contextual Crowdsourced Sensors Things 4

6 Astounding JavaScript * Performance With asm.js Epic * Games Unreal Engine * 3 asm.js LLVM Bitcode JavaScript * web Emscripten Very efficient code generated by Firefox * JIT Over 1M lines of C/C++ code compiled to JavaScript * by Mozilla * and Epic Achieving ~ 1.5x native running time via targeting asm.js, a highly optimizable subset of JavaScript defined by Mozilla asm.js : a highly optimizable low-level subset of JavaScript Courtesy of Mozilla Alon Zakai & Luke Wagner ( 6

7 The March of Parallelism Intel Advanced Vector Extensions 8X peak SIMD operations per core over 4 generations 22 nm Tock AVX-512 AVX2 FMA and integer support 2X 512-bit vectors Next Gen Intel Xeon Phi TM 2X 22 nm Tick AVX 256-bit floating point 2X X=128-bit Since nm Tock Modern processors utilize parallelism to deliver high performance within a constrained power budget 7

8 Optimizing Web Runtimes for Parallelism JavaScript * Parse + build DOM Layout Engine JS Other 11% 11% 21% 21 Layout % 33% 33% GPU: parallel Render 35% 35 % Render CPU: mainly single-threaded HTML5 runtimes of today are not scalable with number of cores Need parallelism for both responsiveness and energy efficiency Web runtimes need to be parallel end-to-end 8

9 Parallel Parsing and Compilation bootstrap 4 threads launch 1 thread PESPMA 2009 JS and GFX execution Four threads for JavaScript * parsing and compilation Epic * Citadel * profile on Firefox * Cycle Breakdown by Categories js::compile gfx::compile os::others js::parse js::others browser::others os::mem js::jitted gfx::exec Background JIT compilers now in Chrome *, Firefox, Internet Explorer *, Safari * 9

10 Towards Parallelizing the Browser Layout Engine Layout Engine: a performance bottleneck Mozilla * Firefox * Page-Load Tests ul em {color:blue} Zimbra * Collaboration Suite * Layout Engine ~42% execution HotPar 2010 CSS rule matching ~33% of the layout Browser layout engine is a bottleneck but amenable to parallelism 10

11 Parallel JavaScript * Started at Intel Labs, now with Mozilla * Extends JavaScript * with a data-parallel API Designed for multi-core CPUs and GPUs Simple, portable, and secure Sequential Array increment example: A.map(function(a) {return a+1;}); A.mapPar(function(a) {return a+1;}); Parallel Accelerated animation of 3D avatars: more characters and more realism Parallel JavaScript goal is to enable data-parallelism in web applications 11

12 SIMD Single Instruction, Multiple Data A x Scalar Operation + B x = C x SIMD Operation of Vector Length 4 A x B x C x A y + B y = C y A y B y + = C y A z + B z = C z A z A w B z B w C z C w A w + B w = C w Intel Architecture currently has SIMD operations of vector length 4, 8, 16 SIMD operations deliver great performance & power efficiency 12

13 SIMD - A Gap Between JavaScript * and Native C++ code for list average SIMD code by ICC A Google * /Intel/Mozilla * ECMA TC39 Joint Project Proposed JavaScript * code Bugzilla * : John McCutchan s strawman proposal: SIMD in JavaScript further reduces the performance gap Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark* and MobileMark*, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. 13

14 SIMD.JS The API Our Firefox * Prototype Our SIMD prototype delivers 3x~4x Mandelbrot speedup Initial support for float32x4 and int32x4 14

15 Demo: Combining SIMD and Higher-Level Parallelism WW: Number of WebWorkers Our Chromium * Prototype SIMD speedup is nicely multiplied by WebWorkers Source: Intel Peter Jensen : 15

16 SIMD Speedups on our Chromium * Prototype SIMD x-times faster than non-simd Theoretical speedup limit is 4 3 rd Generation Intel Core i7 processor (3667U)@ 2.00 GHz, 32-bit, Ubuntu * 13 3 rd Generation Intel Core i7 processor (3667U)@ 2.00 GHz, 64-bit, Ubuntu * 13 Intel Atom processor 1.46GHz, Android * Transpose4x4 AOBench Mandelbrot MatrixMultiplication VertexTransform Average ShiftRows Matrix4x4Inverse Excellent early results while still focused on functionality SIMD.JS benchmarks: 16

17 SIMD.JS Proposal and Polyfill API SIMD Number (Google s John McCutchan & Intel s Peter Jensen): Polyfill API: float32x4, int32x4, Float32x4Array, Int32x4Array Constructors: float32x4(x,y,z,w) float32x4.zero() float32x4.splat(s) Operations: abs, neg, add, sub, mul, div, clamp, min, max, reciprocal, reciprocalsqrt, scale, sqrt, shuffle, shufflemix, withx, withy, withz, withw, lessthan, lessthanorequal, equal, notequal, greaterthanorequal, greaterthan, bitstoint32x4, toint32x4, The joint Google * /Intel/Mozilla * SIMD.JS proposal was approved to advance to the next stage of ECMAScript * TC39 standardization stage A copy of the TC39 Presentation: 17

18 Emscripten now targets SIMD.JS C/C++ JavaScript * Scalar JS Scalar C++ SIMD JS SIMD C Speedup over Scalar JS Near-native SIMD.JS speedup Emscripten generates SIMD.JS from C++ SIMD intrinsics & auto-vectorized code 18

19 Crosswalk in Brief Latest HTML5 features in packaged web apps Today on Android * and Tizen * Based on web technologies: HTML5, CSS3, JavaScript * Easy addition of extensible APIs Application Runtime Open Source, using Blink * & Chromium * Focuses on security, performance and standards compliance Follow us crosswalk-project.org Updated & released to the latest Chromium every 6 weeks Easy access to device APIs Intel platform capabilities 19

20 Intel XDK Cross-platform Development Kit Free at Remote debugging & profiling Develop, debug, profile, and build responsive web & hybrid apps 20

22 Toward Perceptual Computing Close-Range Tracking 2D/ 3D Object Tracking Speech Recognition Gesture Recognition Facial Analysis Devices sense & perceive user actions in a natural & intuitive way Source: Intel Perceptual Computing SDK: 22

23 Reinventing Everyday Usages Learning & Education Immersive Collaboration 3D Scanning and Sharing Scan it Gaming Out-of-reach Device Input Share it Customize & Print it Perceptual Computing opens up new dimensions in interacting with machine 23

24 Demos: Media Capture Depth Stream Extension Source: Intel Ningxin Hu: WebRTC Google * Code: Magic Xylophone: Soundstep *.com: 24

25 Enabling 3D Camera on Web Platform 3D Camera Beyond color: additional per-pixel distance Intel RealSense on PC & tablets soon Applications Real-time hand/finger/object tracking 3D scanning Video conferencing Depth on Web Platform Media Capture Depth Stream Extension Rendering & post-processing: <video>, <canvas>, WebGL * and SIMD.JS Streaming: transmit as MediaStream via WebRTC RTCPeerConnection Source: Intel Ningxin Hu: 25

26 Proposed Media Capture Depth Stream Extension Web Application RGB Stream Depth Stream getusermedia (WebRTC) API Browser or HTML5 runtime Source: 26

27 Wireless Display for the Web Media Sharing/Casting Presentation Gaming Unlock exciting new user experiences in HTML5 Big Buck Bunny video: 27

28 HTML5 Presentation API Proposal Connects web content to screens around you Hides display connection technologies from the developer Apple * AirPlay *, Microsoft * PlayTo *, Google * Chromecast *, Miracast *, Intel Widi Simple, high level API, easy to use New standards-based feature for the cross-platform web Source: Intel Dominik Röttsches 28

30 Intel XDK IoT Edition Companion Apps Streamlined Workflow Design, Test, and Build Tools Quick start samples and templates Built-in editor and emulators UI Frameworks and Apache Cordova* APIs Test and debug tools Integration with Cloud Services APIs Design and build cross-platform companion apps easily for Android*, ios*, and Windows* 30

31 Intel XDK IoT Edition JavaScript* apps on IoT devices Integrated Development Environment Create, Debug, and Run Tools JavaScript allows easy on-board app development and deployment for many IoT devices Use JavaScript to define behavior of IoT device Deploy, run, debug on IoT device with JavaScript Integration with cloud, web services, and sensors through JavaScript APIs Development System Development Platform Edit JavaScript app Send app to device Run app remotely Remote debug IoT Device 31

32 Demo: Programming Internet of Things using Intel XDK IoT Edition Internet of Things (IoT) Device (Intel Galileo): RGB Lighting PWM Led Controller on I2C bus RGB Led Node.js with Socket.io server HTML App (Lenovo * K900): Socket.io connection to IoT device Change lighting color Cordova * App Both made using: Intel XDK IoT Edition Source: Intel Dan Yocom: 32

34 Summary HTML5 is closing the gaps with native models SIMD in JavaScript * enables a large new class of high-performance apps JavaScript is about to get a lot faster for such domains as gaming Depth Camera support in HTML5 WebRTC enables exciting use cases JavaScript is proliferating rapidly in Internet of Things Intel XDK supports end-to-end programming for Internet of Things HTML5 is the application model of the future 34

35 Web: The Ubiquitous Software Platform Rich Capabilities & Content Big Data and the Application Model of the Future Social Contextual Crowdsourced Sensors Things 35

36 Call to Action Download Firefox * Nightly and experience the benefits of SIMD.JS Leverage the power of SIMD.JS through Intel XDK and Crosswalk Download Intel XDK free at SIMD.JS demos: 36

37 Intel Developer Zone Tools. Knowledge. Community. Free tools and code samples Technical articles, forums and tutorials Connect with Intel and industry experts Get development support Build relationships software.intel.com 37

38 Legal Disclaimer INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. A "Mission Critical Application" is any application in which failure of the Intel Product could result, directly or indirectly, in personal injury or death. SHOULD YOU PURCHASE OR USE INTEL'S PRODUCTS FOR ANY SUCH MISSION CRITICAL APPLICATION, YOU SHALL INDEMNIFY AND HOLD INTEL AND ITS SUBSIDIARIES, SUBCONTRACTORS AND AFFILIATES, AND THE DIRECTORS, OFFICERS, AND EMPLOYEES OF EACH, HARMLESS AGAINST ALL CLAIMS COSTS, DAMAGES, AND EXPENSES AND REASONABLE ATTORNEYS' FEES ARISING OUT OF, DIRECTLY OR INDIRECTLY, ANY CLAIM OF PRODUCT LIABILITY, PERSONAL INJURY, OR DEATH ARISING IN ANY WAY OUT OF SUCH MISSION CRITICAL APPLICATION, WHETHER OR NOT INTEL OR ITS SUBCONTRACTOR WAS NEGLIGENT IN THE DESIGN, MANUFACTURE, OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS. Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined". Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design with this information. The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request. Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order. Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling , or go to: Intel, Core, Atom, Xeon Phi, RealSense, Look Inside and the Intel logo are trademarks of Intel Corporation in the United States and other countries. *Other names and brands may be claimed as the property of others. Copyright 2014 Intel Corporation. 38

39