DOCSLIB.ORG

Webcore: Architectural Support for Mobile Web Browsing

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Webcore: Architectural Support for Mobile Web Browsing WebCore: Architectural Support for Mobile Web Browsing Yuhao Zhu Vijay Janapa Reddi Department of Electrical and Computer Engineering The University of Texas at Austin [email protected], [email protected] Abstract The Web browser is undoubtedly the single most impor- Browser Browser tant application in the mobile ecosystem. An average user 63% 54% spends 72 minutes each day using the mobile Web browser. Web browser internal engines (e.g., WebKit) are also growing 23% 8% 32% Media 6% in importance because they provide a common substrate for 7% 7% Others developing various mobile Web applications. In a user-driven, Media Games Others interactive, and latency-sensitive environment, the browser’s Email performance is crucial. However, the battery-constrained na- (a) Time dist. of window focus. (b) Time dist. of CPU processing. ture of mobile devices limits the performance that we can de- Fig. 1: Mobile Web browser share study conducted by our industry liver for mobile Web browsing. As traditional general-purpose research partner on their employees’ devices [2]. Similar observa- techniques to improve performance and energy efficiency fall tions were reported by NVIDIA on Tegra-based mobile handsets [3,4]. short, we must employ domain-specific knowledge while still maintaining general-purpose flexibility. network limited. However, this trend is changing. With about In this paper, we first perform design-space exploration 10X improvement in round-trip time from 3G to LTE, network to identify appropriate general-purpose architectures that latency is no longer the only performance bottleneck [51]. uniquely fit the characteristics of a popular Web browsing Prior work has shown that over the past decade, network engine. Despite our best effort, we discover sources of energy technology advancements have managed to keep webpage inefficiency in these customized general-purpose architectures. transmission overhead almost stable, whereas the client-side To mitigate these inefficiencies, we propose, synthesize, and computational requirements have increased by as much as evaluate two new domain-specific specializations, called the 10X [78]. Fig. 1b confirms that the browser consumes a sig- Style Resolution Unit and the Browser Engine Cache. Our opti- nificant portion of the CPU time. Similarly, other research mizations boost energy efficiency and at the same time improve reports over 80% CPU usage for mobile browsing [51]. mobile Web browsing performance. As emerging mobile work- Providing high processing capability in mobile CPUs is loads increasingly rely more on Web browser technologies, the challenging due to the limited battery capacity. Recent studies type of optimizations we propose will become important in the suggested that the advancement in Lithium-ion battery den- future and are likely to have lasting widespread impact. sity has slowed down significantly, and that battery capacity will be mostly limited by its volume [6, 72]. Under such a 1. Introduction constraint, we are now faced with a challenge. On one hand, The proliferation of mobile devices and the fast penetration the energy budget for mobile devices is unlikely to drastically of new Web technologies (such as HTML5) has ushered in a increase in the short term. On the other hand, mobile proces- new era of mobile Web browsing. Mobile users now spend sors are becoming power hungry. Core designs have not only a significant amount of time on the Web browser every day. gone from in-order to out-of-order (e.g., ARM Cortex-A8 to A study conducted by our industry partner reports that the A15), but they have also gone multicore (e.g., Exynos 5410 in browser occupies 63% of the window focus time on their Samsung Galaxy S4 has eight cores). Experiments conducted employees’ mobile devices, as shown in Fig. 1a. As a general by Carroll et al. suggested that processor power of mobile trend, comScore shows that mobile users prefer Web browsers Web browsing doubled from 2010 to 2013 [38, 39]. over native applications for important application domains Our work aspires to bridge the widening gap between high- such as online e-commerce and electronic news feeds [1]. performance and energy-constrained mobile processor designs In the user-centric mobile Web browsing context, delivering for Web browsing. Domain-specific specializations have long high performance is critical to end users’ quality-of-service been known to be extremely energy efficient [57,71,76]. Re- experience. Studies in 2013 indicate that 71% of mobile Web cent proposals in data computation domains (such as H.264 en- users expect website performance on their mobile devices to coding [49] and convolution [67]) have begun showing that it be not worse than their desktop experience–up from 58% in is critical and feasible to balance the efficiency of application- 2009 [5]. Traditionally, Web browsing performance has been specific specializations with general-purpose programmability. 978-1-4799-4394-4/14/$31.00 c 2014 IEEE User JavaScript DOM DOM Actions Engine 17% Layout 16% Render Layout Render 21% 20% Data structures 11% CSS 10% DOM Style Tree Kernels Rules 19% Style 28% 24% GPU Style 35% Others Servers Dom Style Layout Render Painting Others Render Tree Fig. 3: Execution time break- Fig. 4: Energy consumption Browser Engine (our focus) down of the browser’s kernels. breakdown of the kernels. Fig. 2: Web browser overview. 2. Web Browser Background and Overview Sharing the same architecture design philosophy, we pro- We first describe the Web browser engine’s computation ker- pose the WebCore, a general-purpose core customized and nels and their communication patterns. Such understanding specialized for the mobile Web browsing workload. In compar- helps us design effective customizations and specializations, ison to prior work that either takes a fully software approach such as those described in the later sections. Nearly all main- on general-purpose processors [41, 62] or a fully hardware stream browser engines fit into our description. In addition, specialization approach [34], our design strikes a balance be- we show each kernel’s performance and energy breakdown to tween the two. On one hand, WebCore retains the flexibility show their importance and demonstrate our study’s coverage. and programmability of a general-purpose core. It naturally Browser engine kernels Fig. 2 shows the overall flow fits in the multicore SoC that is already common in today’s of execution within any typical Web browser. The engine mainstream mobile devices. On the other hand, it achieves has two core modules: the browser engine (e.g., WebKit for energy-efficiency improvement via modest hardware special- Chrome and Gecko for Firefox) and the JavaScript engine. izations that create closely coupled datapath and data storage. In this work, we focus on the internals of the Web browser We begin by examining existing general purpose designs engine. The JavaScript engine’s performance that involves the for the mobile Web browsing workload. Through exhaustive compiler, garbage collector, etc., is a separate issue beyond the design space exploration, we find that existing general pur- scope of this work. Please refer to the Related Work section pose designs bear inherent sources of energy-inefficiency. In (Sec. 8) for a more elaborate discussion about JavaScript. particular, instruction delivery and data feeding are two major The browser engine mainly consists of four kernels: Dom, bottlenecks. We show that customizing them by tuning key Style, Layout, and Render. The kernels, shown in boxes, pro- design parameters achieves better energy efficiency (Sec. 4). cess the webpage and prepare pixels for a GPU to paint. The Building on the customized general-purpose baseline, we figure also shows the important data structures that the kernels develop specialized hardware to further overcome the instruc- consume. The DOM tree, CSS style rules, and Render tree are tion delivery and data feeding bottlenecks (Sec. 5). We propose those important data structures, and they are heavily shared two new optimizations: the “Style Resolution Unit” (SRU) and across the kernels. The data structures are shown in circles a “Software-Managed Browser Engine Cache.” The SRU is with arrows indicating information flow between the kernels. a software-assisted hardware accelerator for the critical style- The Dom kernel is in charge of parsing the webpage con- resolution kernel within the Web browser engine. It exploits tents. Specifically, it constructs the DOM tree from the HTML fine-grained parallelism that aggregates enough computation files, and extracts the CSS style rules from the CSS files. Given to offset the instruction and data communication overhead. the DOM tree and CSS style rules, the Style kernel computes The proposed cache structure exploits the unique data locality the webpage’s style information and stores the results in the of the browser engine’s principal data structures. It is a small render tree. Each render tree node corresponds to a visible and fast memory that achieves a high hit rate for the important element in the webpage. Once the style information of each data structures, but with extremely low accessing energy. webpage element is calculated, the Layout kernel recursively Our results show that customizations alone on the exist- traverses the render tree to decide each visible element’s po- ing general-purpose mobile processor design lead to 22.2% sition information based on each element’s size and relative performance improvement and 18.6% energy saving. Our spe- positioning. The final < x;y > coordinates are stored back cialization techniques achieve an additional 9.2% performance into the render tree. Eventually, the Render kernel examines improvement and 22.2% overall energy saving; the acceler- the render tree to decide the z-ordering of each visible element ated portion itself achieves up to 10X speedup. Finally, we so that they can be displayed in the correct overlapping order. also show that our specialization incurs negligible area over- Performance Fig.
Load more

Recommended publications
  • EVA: an Efficient Vision Architecture for Mobile Systems
    EVA: An Efficient Vision Architecture for Mobile Systems Jason Clemons, Andrea Pellegrini, Silvio Savarese, and Todd Austin Department of Electrical Engineering and Computer Science University of Michigan Ann Arbor, Michigan 48109 fjclemons, apellegrini, silvio, [email protected] Abstract The capabilities of mobile devices have been increasing at a momen- tous rate. As better processors have merged with capable cameras in mobile systems, the number of computer vision applications has grown rapidly. However, the computational and energy constraints of mobile devices have forced computer vision application devel- opers to sacrifice accuracy for the sake of meeting timing demands. To increase the computational performance of mobile systems we Figure 1: Computer Vision Example The figure shows a sock present EVA. EVA is an application-specific heterogeneous multi- monkey where a computer vision application has recognized its face. core having a mix of computationally powerful cores with energy The algorithm would utilize features such as corners and use their efficient cores. Each core of EVA has computation and memory ar- geometric relationship to accomplish this. chitectural enhancements tailored to the application traits of vision Watts over 250 mm2 of silicon, typical mobile processors are limited codes. Using a computer vision benchmarking suite, we evaluate 2 the efficiency and performance of a wide range of EVA designs. We to a few Watts with typically 5 mm of silicon [4] [22]. show that EVA can provide speedups of over 9x that of an embedded To meet the limited computation capability of mobile proces- processor while reducing energy demands by as much as 3x. sors, computer vision application developers reluctantly sacrifice image resolution, computational precision or application capabili- Categories and Subject Descriptors C.1.4 [Parallel Architec- ties for lower quality versions of vision algorithms.
    [Show full text]
  • 6Th Generation Intel® Core™ Processors Based on the Mobile U-Processor for Iot Solutions (Intel® Core™ I7-6600U, I5-6300U, and I3-6100U Processors)
    PLATFORM BRIEF 6th Generation Intel® Core™ Mobile Processor Family Internet of Things 6th Generation Intel® Core™ Processors Based on the Mobile U-Processor for IoT Solutions (Intel® Core™ i7-6600U, i5-6300U, and i3-6100U Processors) Harness the Performance, Features, and Edge-to-Cloud Scalability to Build Tomorrow’s IoT Solutions Today Product Overview Stunning Visual Performance Intel is proud to announce its 6th The 6th generation Intel Core generation Intel® Core™ processor processors utilize the new Gen9 family featuring ultra low-power, graphics engine, which improves 64-bit, multicore processors built on graphic performance by up to the latest 14 nm technology. Designed 34 percent.1 The improvements are for small form-factor applications, this demonstrated through faster 3-D multichip package (MCP) integrates graphics performance and rendering a low-power CPU and platform applications at low power. Video controller hub (PCH) onto a common playback is also faster and smoother package substrate. thanks to the new multiplane overlay capability. The new generation offers The 6th generation Intel Core processor up to three independent audio streams family offers dramatically higher CPU and displays, Ultra HD 4K support, and and graphics performance, a broad workload consolidation for lower BOM range of power and features scaling costs and energy output. the entire Intel product line, and new, advanced features that boost edge-to- Users will also enjoy enhanced cloud Internet of Things (IoT) designs high-density streaming applications in a wide variety of markets. These and optimized 4K videoconferencing processors run at 15W thermal design with accelerated 4K hardware media power (TDP) and are ideal for small, codecs HEVC (8-bit), VP8, VP9, and energy-efficient, form-factor designs, VDENC encoding, decoding, and including digital signage, point-of-sale transcoding.
    [Show full text]
  • PART I ITEM 1. BUSINESS Industry We Are
    PART I ITEM 1. BUSINESS Industry We are the world’s largest semiconductor chip maker, based on revenue. We develop advanced integrated digital technology products, primarily integrated circuits, for industries such as computing and communications. Integrated circuits are semiconductor chips etched with interconnected electronic switches. We also develop platforms, which we define as integrated suites of digital computing technologies that are designed and configured to work together to provide an optimized user computing solution compared to ingredients that are used separately. Our goal is to be the preeminent provider of semiconductor chips and platforms for the worldwide digital economy. We offer products at various levels of integration, allowing our customers flexibility to create advanced computing and communications systems and products. We were incorporated in California in 1968 and reincorporated in Delaware in 1989. Our Internet address is www.intel.com. On this web site, we publish voluntary reports, which we update annually, outlining our performance with respect to corporate responsibility, including environmental, health, and safety compliance. On our Investor Relations web site, located at www.intc.com, we post the following filings as soon as reasonably practicable after they are electronically filed with, or furnished to, the U.S. Securities and Exchange Commission (SEC): our annual, quarterly, and current reports on Forms 10-K, 10-Q, and 8-K; our proxy statements; and any amendments to those reports or statements. All such filings are available on our Investor Relations web site free of charge. The SEC also maintains a web site (www.sec.gov) that contains reports, proxy and information statements, and other information regarding issuers that file electronically with the SEC.
    [Show full text]
  • NVIDIA Tegra 4 Family CPU Architecture 4-PLUS-1 Quad Core
    Whitepaper NVIDIA Tegra 4 Family CPU Architecture 4-PLUS-1 Quad core 1 Table of Contents ...................................................................................................................................................................... 1 Introduction .............................................................................................................................................. 3 NVIDIA Tegra 4 Family of Mobile Processors ............................................................................................ 3 Benchmarking CPU Performance .............................................................................................................. 4 Tegra 4 Family CPUs Architected for High Performance and Power Efficiency ......................................... 6 Wider Issue Execution Units for Higher Throughput ............................................................................ 6 Better Memory Level Parallelism from a Larger Instruction Window for Out-of-Order Execution ...... 7 Fast Load-To-Use Logic allows larger L1 Data Cache ............................................................................. 8 Enhanced branch prediction for higher efficiency .............................................................................. 10 Advanced Prefetcher for higher MLP and lower latency .................................................................... 10 Large Unified L2 Cache .......................................................................................................................
    [Show full text]
  • Architectural Support for Javascript Type-Checking on Mobile Processors
    Checked Load: Architectural Support for JavaScript Type-Checking on Mobile Processors Owen Anderson Emily Fortuna Luis Ceze Susan Eggers Computer Science and Engineering, University of Washington http://sampa.cs.washington.edu Abstract applications, including Google Maps, Twitter, and Face- book would not be feasible without both high-throughput Dynamic languages such as Javascript are the de-facto and low-latency JavaScript virtual machines on the client. standard for web applications. However, generating effi- At the same time, innovations in mobile device pro- cient code for dynamically-typed languages is a challenge, grammability have opened up embedded targets to the same because it requires frequent dynamic type checks. Our anal- class of programmers. Today’s smart mobile devices are ysis has shown that some programs spend upwards of 20% expected to provide a developer API that is usable by of dynamic instructions doing type checks, and 12.9% on normal application developers, as opposed to the special- average. ized embedded developers of the past. One such plat- In this paper we propose Checked Load, a low- form, HP/Palm’s WebOS [17], uses JavaScript as its pri- complexity architectural extension that replaces software- mary application development language. Others encourage based, dynamic type checking. Checked Load is comprised JavaScript-heavy web applications in addition to their na- of four new ISA instructions that provide flexible and au- tive development environments, as a means of providing tomatic type checks for memory operations, and whose im- feature-rich, portable applications with minimal develop- plementation requires minimal hardware changes. We also ment costs. propose hardware support for dynamic type prediction to Because of their power and space constraints, embedded reduce the cost of failed type checks.
    [Show full text]
  • Intel® Core™2 Duo Mobile Processor for Intel® Centrino® Duo Mobile Processor Technology
    Intel® Core™2 Duo Mobile Processor for Intel® Centrino® Duo Mobile Processor Technology Datasheet September 2007 Document Number: 314078-004 INFORMATIONLegal Lines and Disclaimers 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. UNLESS OTHERWISE AGREED IN WRITING BY INTEL, THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A SITUATION WHERE PERSONAL INJURY OR DEATH MAY OCCUR. 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.
    [Show full text]
  • Design and Implementation of Low Power 5 Stage Pipelined 32 Bits MIPS Processor Using 28Nm Technology
    International Journal of Innovative Technology and Exploring Engineering (IJITEE) ISSN: 2278-3075, Volume-8 Issue-4S2 March, 20 Design and implementation of low power 5 stage Pipelined 32 bits MIPS Processor using 28nm Technology V.Prasanth, V.Sailaja, P.Sunitha, B.Vasantha Lakshmi processor. The operations used by MIPS processor in Abstract— MIPS is a simple streamlined highly scalable instruction set which are generally used to access memory in RISC architecture is most used in android base devices and best MIPS processor are load and Store and other operations suited for portable mobile devices. This Paper presents a design which are remaining are performed on register to register of 5 stage pipelined 32 bit MIPS processor on a 28nm basis [2] this results in more clear instruction set design Technology. The processor is designed using Harvard architecture. The most important feature of pipelining is where it allows execution of one instruction-per cycle rate. performance and speed of the processor, this results in increase The pipelining uses parallelism at instruction level to execute of device power. To reduce dynamic power using RTL clock multiple instructions simultaneously using a single processor gating inside FPGA device we presented a novel approach in this [2]. The major disadvantage with MIPS processor design is paper. Design functionality in terms of area power and speed is dynamic power consumption results due to clock power and analyzed using kintex 7 platform board. switching-activity. Clock Gating is a method which employed in the design to reduce power consumption [17] by Keywords: RISC,MIPS, Clock Gating, Dynamic Power, reducing switching activity of non active blocks.
    [Show full text]
  • 5 Microprocessors
    Color profile: Disabled Composite Default screen BaseTech / Mike Meyers’ CompTIA A+ Guide to Managing and Troubleshooting PCs / Mike Meyers / 380-8 / Chapter 5 5 Microprocessors “MEGAHERTZ: This is a really, really big hertz.” —DAVE BARRY In this chapter, you will learn or all practical purposes, the terms microprocessor and central processing how to Funit (CPU) mean the same thing: it’s that big chip inside your computer ■ Identify the core components of a that many people often describe as the brain of the system. You know that CPU CPU makers name their microprocessors in a fashion similar to the automobile ■ Describe the relationship of CPUs and memory industry: CPU names get a make and a model, such as Intel Core i7 or AMD ■ Explain the varieties of modern Phenom II X4. But what’s happening inside the CPU to make it able to do the CPUs amazing things asked of it every time you step up to the keyboard? ■ Install and upgrade CPUs 124 P:\010Comp\BaseTech\380-8\ch05.vp Friday, December 18, 2009 4:59:24 PM Color profile: Disabled Composite Default screen BaseTech / Mike Meyers’ CompTIA A+ Guide to Managing and Troubleshooting PCs / Mike Meyers / 380-8 / Chapter 5 Historical/Conceptual ■ CPU Core Components Although the computer might seem to act quite intelligently, comparing the CPU to a human brain hugely overstates its capabilities. A CPU functions more like a very powerful calculator than like a brain—but, oh, what a cal- culator! Today’s CPUs add, subtract, multiply, divide, and move billions of numbers per second.
    [Show full text]
  • Introducing Intel's First-Ever Core™ I9 Mobile Processor
    Product Brief Mobile 8TH Gen Intel® Core™ Processors INTRODUCING INTEL’s FIRST-EVER CORE™ I9 MOBILE PROCESSOR. THE PERFORMANCE POWERHOUSE FOR WHAT THE MOST DEMANDING MOBILE ENTHUSIASTS NEED TODAY AND FOR WHAT COMES NEXT. Be ready for amazing experiences in gaming, VR, content creation, and entertainment wherever your computing takes you with the highest- performance mobile 8th Generation Intel® Core™ processor family. This latest addition to the 8th generation processor family extends all the capabilities users have come to love in our mobile H series platforms with advanced innovations that deliver exciting new features to immerse you in incredible experiences on a variety of form factors. Product Brief Mobile 8TH Gen Intel® Core™ Processors REDEFINE ENTHUSIAST MOBILE PC PERFORMANCE ULTIMATE MOBILE PLATFORM PERFORMANCE The newest 8th Generation Intel Core processors redefine enthusiast mobile PC performance now with up to six cores and 12 MB of cache memory for more processing power—that’s two more cores than the previous generation Intel Core processor family—Intel® Turbo Boost 2.0 technology and new Intel® Thermal Velocity Boost to opportunistically and automatically increase core frequency whenever processor temperature and turbo budget allows. Intel® Hyper-Threading Technology delivers multitasking support in the latest generation of Intel Core processors. For the enthusiast, the fully-unlocked 8th Generation Intel® Core™ i9-8950HK processor provides the opportunity to tweak the platform performance to its fullest potential and enjoy hardcore mobile gaming and VR experiences. 2 Product Brief Mobile 8TH Gen Intel® Core™ Processors THE NEW MOBILE 8TH GENERATION INTEL® CORE™ PROCESSOR FAMILY DELIVERS: • An impressive portfolio of standard and unlocked systems for a broad range of usages and performance levels.
    [Show full text]
  • Intel® Core™ I7-900/I7-800/I7-700 Mobile Processor
    Intel® Core™ i7-900 Mobile Processor Extreme Edition Series, Intel Core i7-800 and i7-700 Mobile Processor Series Datasheet- Volume One This is volume 1 of 2. Refer to document 320766 for Volume 2 September 2009 Document Number: 320765 -001 INFORMATIONLegal Lines and Disclaimers 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. UNLESS OTHERWISE AGREED IN WRITING BY INTEL, THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A SITUATION WHERE PERSONAL INJURY OR DEATH MAY OCCUR. 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.
    [Show full text]
  • Android in the Cloud on Arm Native Servers
    Android in the cloud on Arm native servers December 2020 Executive summary All the chatter about Arm servers and the role of Android in the cloud in recent years generates questions about what these things could mean for a largely mobile based platform like Arm/Android. This whitepaper examines the practical and commercial use cases that are driving a new paradigm linking the primary mobile and embedded computing platform in the world today with its new cousin in cloud infrastructure: Arm servers + virtualisation + native Android execution. While detailing the use cases driving a mobile platform migration to cloud infrastructure, this whitepaper also showcases the solution that Ampere, Canonical, and NETINT Technologies have partnered to produce addressing this migration. The solution is an example of how a service provider or developer can take advantage of Arm native computing in a cloud context to bring together an ecosystem of over 3 million primarily mobile apps integrated with benefits from a cloud enabled infrastructure. A case study around cloud gaming fills in the practical real-world considerations of this new paradigm. Cloud gaming really showcases the advantages of all the solution components the partners have integrated to make a scalable, efficient cloud resource available to a new class of cloud-native applications catering to the billions of users with Arm based devices worldwide. The whitepaper concludes with thoughts and predictions about the evolution of this new paradigm as hardware and software components mature in the coming years. A new paradigm The Arm architecture has dominated the mobile processor market with its unrivaled ability to maximise power-efficiency.
    [Show full text]
  • Cubesat and Mobile Processors
    CubeSat and Mobile Processors Steven M. Guertin [email protected] 818-321-5337 NASA/JPL Acknowledgment: This work was sponsored by: The NASA Electronic Parts and Packaging Program (NEPP) To be presented by Steven M. Guertin at the NASA Electronics Technology Workshop, June 23-26, 2015. 1 Outline • Processor/Microcontroller Review • Collaborations • MSP430 • PIC • Atmel AT91SAM9G20 • Intel Atom E620 • Snapdragon (APQ8064) • Future/Conclusions To be presented by Steven M. Guertin at the NASA Electronics Technology Workshop, June 23-26, 2015. 2 Motivation • Provide SEE and TID test data on processors and microcontrollers of interest for CubeSat and Small Missions – CubeSat Kit devices – Devices either flying or in designs being built – (and if enough interest, devices people would like to fly) – Future-looking devices • Microcontroller focus is primarily on devices that are in designs right now – these are essentially the embedded market • Microprocessor focus is on mobile devices where power is low and performance to power ratio is high To be presented by Steven M. Guertin at the NASA Electronics Technology Workshop, June 23-26, 2015. 3 Microcontroller Review • Review of CubeSat kits • Review of NASA CubeSat parts lists • Direct interaction with community members To be presented by Steven M. Guertin at the NASA Electronics Technology Workshop, June 23-26, 2015. 4 Mobile Processor Review • Various CubeSats have flown with more capable processing – Usually C&DH is 8 or 16-bit MCUs, the “ARMs” are actually reduced capability Thumb™ processors – AAUSat-3, CANX-2 used a 32-bit ARM processor – Phonesat … flew … phones (and newer iterations are flying more) • Expect that as CubeSat programs continue, need for more processing will be important • Key drivers – same as for CubeSats in general – Small, low power, cheap – Generally accessible to low-budget environments (schools, R&D, etc.) To be presented by Steven M.
    [Show full text]