Academia.eduAcademia.edu

Outline

Title
Abstract
FAQs
Figures
Introduction
Os Design Space Exploration
Design Objective
Methodology
Analysis of Engineering Efforts
Energy Benefits Methodology
Discussions
Related Work
Conclusions
References

Transkernel: An Executor for Commodity Kernels on Peripheral Cores

Profile image of Nurhan AkarasNurhan Akaras

2018

https://doi.org/10.25394/PGS.8277515.V1
visibility

…

description

17 pages

link

1 file

Abstract

Modern mobile devices have numerous ephemeral tasks. These tasks are driven by background activities, such as push notifications and sensor readings. In order to execute these tasks, the whole platform has to periodically wake up beforehand, and go to sleep afterwards. During this process, the OS kernel operates on power state of various IO devices, which has been identified as the bottleneck for energy efficiency. To this end, we want to offload this kernel phase to a more energy efficient, microcontroller level core, named peripheral core. To execute commodity OS on a peripheral core, existing approaches either require much engineering effort or incur high execution cost. Therefore, we proposed a new OS model called transkernel. By utilizing cross-ISA dynamic binary translation (DBT) technique, transkernel creates a virtualized environment on the peripheral core. It relies on a small set of stable interfaces. It is specialized for frequently executed kernel path. It exploits ISA s...

FAQs

sparkles

AI

How does transkernel improve energy efficiency during device suspend/resume phases?add

Transkernel reduces system energy consumption by 34% during device suspend/resume phases, significantly outperforming native execution methods.

What are the primary challenges in offloading kernels to peripheral cores?add

Key challenges include managing different ISAs, ensuring binary compatibility during kernel evolution, and maintaining effective communication protocols.

How does the ARK implementation handle interrupts during execution?add

ARK emulates a brief stage of interrupt handling for ISA-specific tasks before transitioning to the kernel's neutral handling routine.

What criteria define the stable ABI utilized by transkernel for compatibility?add

The stable ABI consists of 12 Linux kernel functions and a variable that have remained unchanged for several Linux versions since 2014.

What execution overhead does ARK incur compared to native kernel execution?add

ARK incurs an execution overhead of 2.7× on average, which is significantly lower than the 13.9× overhead seen in baseline DBT implementations.

Figures (10)
Figure 1: An overview of this work drives the whole hardware platform out of deep sleep before- hand (i.e., “resume’’) and puts it back to deep sleep afterwards (i.e., “suspend’’). During this process, the kernel consumes much more energy than the user code [44], up to 10x shown in recent work [38].
Figure 1: An overview of this work drives the whole hardware platform out of deep sleep before- hand (i.e., “resume’’) and puts it back to deep sleep afterwards (i.e., “suspend’’). During this process, the kernel consumes much more energy than the user code [44], up to 10x shown in recent work [38].
Table 1: Our hardware model fits many popular SoCs which are used in popular products such as Apple Watch and Azure Sphere. Section 7.5 discusses caveats. *: lack public technical details.
Table 1: Our hardware model fits many popular SoCs which are used in popular products such as Apple Watch and Azure Sphere. Section 7.5 discusses caveats. *: lack public technical details.
(a) # of functions (b) # of functions (upper) & types (lower) w/ changed ABI across kernel versions Figure 2: Alternative ways for offloading kernel phases
(a) # of functions (b) # of functions (upper) & types (lower) w/ changed ABI across kernel versions Figure 2: Alternative ways for offloading kernel phases
Figure 4: The ARK structure on a peripheral core
Figure 4: The ARK structure on a peripheral core
*=ABI unchanged since 2014 (v3.16); others unchanged since 2011 (v2.6).
*=ABI unchanged since 2014 (v3.16); others unchanged since 2011 (v2.6).
Figure 5: Execution time and energy in device suspend/re- sume. ARK substantially reduces the energy.
Figure 5: Execution time and energy in device suspend/re- sume. ARK substantially reduces the energy.
Figure 6: Busy execution overhead for devices under test (top: suspend; bottom: resume). Our DBT optimizations reduce the overhead by up to one order of magnitude
Figure 6: Busy execution overhead for devices under test (top: suspend; bottom: resume). Our DBT optimizations reduce the overhead by up to one order of magnitude
Figure 7: System energy consumption (inc. cores, DRAM, and IO) of ARK relative to native execution (100%), under different DBT overheads (x-axis) and processor core usage (y-axis). ARK’s low energy hinges on low DBT overhead.
Figure 7: System energy consumption (inc. cores, DRAM, and IO) of ARK relative to native execution (100%), under different DBT overheads (x-axis) and processor core usage (y-axis). ARK’s low energy hinges on low DBT overhead.

Related papers

Spumone: Lightweight cpu virtualization layer for embedded systems
Tatsuo Nakajima

2008

Abstract Recently, the engineering cost of embedded systems is rapidly increasing due to growing sophistication of services. To deal with the problem, hybrid operating system environments have been proposed. This enables to run a RTOS and a general purpose OS concurrently and to reuse software resources on both of them. This approach is efficient in reducing engineering costs.

View PDFchevron_right
Power Aware Instruction Scheduling for Microcontrollers
D. Elkaduwe, K. Samarasinghe

Abstract-The ability to reduce power consumption of a device is attractive for several reasons. On one hand, reducing power in high end computers will reduce the cost of cooling and the performance loss due to overheating cores. On the other hand, ubiquitous battery powered devices will enjoy a longer battery life due to the reduction in power consumption. In this paper, we present a set of mechanisms that uses instruction scheduling to reduce the power consumption of RISC like microcontrollers that are common in battery powered ...

View PDFchevron_right
Energy management for real-time embedded applications with compiler support
Rami Melhem

Proceedings of the 2003 ACM SIGPLAN conference on Language, compiler, and tool for embedded systems - LCTES '03, 2003

Reducing device energy has become one of the most important challenges to embedded systems designers. Processors with dynamic voltage scaling permit trading performance for reduced energy consumption as a program executes. In this paper, we first present a novel hybrid scheme that uses dynamic voltage scaling to adjust the performance of embedded applications to reduce energy consumption while also meeting time constraints. Our fine-grained approach uses the compiler to insert power management hints in the application code. These hints convey path-specific runtime information about the program's progress to power management points invoked by the operating system that adjust processor performance. Second, we present an algorithm for inserting power management hints along different program paths. Finally, we experimentally evaluate our approach and show that significant energy reduction can be achieved. On two embedded applications, MPEG movie decoding and automatic target recognition, our scheme reduces energy by up to 79% over no power management and by up to 50% over static power management. We also experimentally demonstrate that our scheme achieves more energy savings compared to two purely compiler-directed schemes.

View PDFchevron_right
The performance and energy consumption of embedded real-time operating systems
Kathleen Baynes, Tiebing Zhang

IEEE Transactions on Computers, 2003

This paper presents the modeling of embedded systems with SimBed, an execution-driven simulation testbed that measures the execution behavior and power consumption of embedded applications and RTOSs by executing them on an accurate architectural model of a microcontroller with simulated real-time stimuli. We briefly describe the simulation environment and present a study that compares three RTOSs: C/OS-II, a popular public-domain embedded real-time operating system; Echidna, a sophisticated, industrial-strength (commercial) RTOS; and NOS, a bare-bones multirate task scheduler reminiscent of typical "roll-your-own" RTOSs found in many commercial embedded systems. The microcontroller simulated in this study is the Motorola M-CORE processor: a low-power, 32-bit CPU core with 16-bit instructions, running at 20MHz. Our simulations show what happens when RTOSs are pushed beyond their limits and they depict situations in which unexpected interrupts or unaccounted-for task invocations disrupt timing, even when the CPU is lightly loaded. In general, there appears no clear winner in timing accuracy between preemptive systems and cooperative systems. The power-consumption measurements show that RTOS overhead is a factor of two to four higher than it needs to be, compared to the energy consumption of the minimal scheduler. In addition, poorly designed idle loops can cause the system to double its energy consumption-energy that could be saved by a simple hardware sleep mechanism.

View PDFchevron_right
Energy-aware task mapping onto heterogeneous platforms using DVFS and sleep states
Patrick Meumeu Yomsi

Real-Time Systems, 2015

Heterogeneous multicore platforms are becoming an interesting alternative for embedded computing systems with limited power supply as they can execute specific tasks in an efficient manner. Nonetheless, one of the main challenges of such platforms consists of optimising the energy consumption in the presence of temporal constraints. This paper addresses the problem of task-to-core allocation onto heterogeneous multicore platforms such that the overall energy consumption of the system is minimised. To this end, we propose a two-phase approach that considers both dynamic and leakage energy consumption: (i) the first phase allocates tasks to the cores such that the dynamic energy consumption is reduced; (ii) the second phase refines the allocation performed in the first phase in order to achieve better sleep states by trading off the dynamic energy consumption with the reduction in leakage energy consumption. This hybrid approach considers core frequency set-points, tasks energy consumption and sleep states of the cores to reduce the energy consumption of the system. Major value has been placed on a realistic power model which increases the practical relevance of the proposed approach. Finally, extensive simulations have been carried out to demonstrate the effectiveness of the proposed algorithm. In the best-case, savings up to 18% of energy are reached over the first fit algorithm, which has shown, in previous works, to perform better than other bin-packing heuristics for the target heterogeneous multicore platform.

View PDFchevron_right
Task Migration for Dynamic Power and Performance Characteristics on Many-Core Distributed Operating Systems
Sunny Khan

Spatial locality of task execution will become more important on future hardware platforms since the number of cores are steadily increasing. The large amount of cores requires more intelligent power management due to the notion of spatial locality, and the high chip density requires an increased thermal awareness in order to avoid thermal hotspots on the chip. At the same time, high performance of the CPU is only achieved by parallelizing tasks over the chip in order to fully utilize the hardware. This paper presents a task migration mechanism for distributed operating systems running on manycore platforms. In this work, we evaluate the performance and energy efficiency of an implemented task migration mechanism. This is shown by parallelizing tasks as the performance of a single core is not sufficient, and by collecting tasks to as few cores as possible as CPU load is low. The task migration mechanism is implemented as a library for FreeRTOS using 1300 lines of code, and introduced a total task migration overhead of 100 ms on a shared memory platform. With the presented task migration mechanism, we intend to improve the dynamism of power and performance characteristics in distributed many-core operating systems.

View PDFchevron_right
Collaborative operating system and compiler power management for real-time applications
Rami Melhem

ACM Transactions on Embedded Computing Systems, 2006

Managing energy consumption has become vitally important to battery-operated portable and embedded systems. Dynamic voltage scaling (DVS) reduces the processor's dynamic power consumption quadratically at the expense of linearly decreasing the performance. When reducing energy with DVS for real-time systems, one must consider the performance penalty to ensure that deadlines can be met. In this paper, we introduce a novel collaborative approach between the compiler and the operating system (OS) to reduce energy consumption. We use the compiler to annotate an application's source code with path-dependent information called power-management hints (PMHs). This fine-grained information captures the temporal behavior of the application, which varies by executing different paths. During program execution, the OS periodically changes the processor's frequency and voltage based on the temporal information provided by the PMHs. These speed adaptation points are called power-manage...

View PDFchevron_right
Guest Editorial: Embedded Multicore Systems and Applications
Kuan Ching Li

Journal of Signal Processing Systems, 2015

View PDFchevron_right
An energy-aware synthesis methodology for OS-driven multi-process embedded software
TAT TAN

2004

The growing software content in various battery-driven embedded systems has led to significant interest in technologies for energy-efficient embedded software. While lowenergy software research has, in the past, focused on energy optimization at the instruction and source-code levels, approaches targeted at a higher software level are beginning to gain attention. In this work, we propose a methodology to refine a control-data flow diagram (CDFD) model of an embedded software program into an energy-efficient multi-process software architecture graph (SAG). Our starting representation, the CDFD, is capable of modeling data-dependent control flow. Energy efficiency is achieved by reducing the energy wastage due to context switches and inter-process communications (IPCs). Conditionally-unused computational operations, and their corresponding energy, are also avoided by a condition-aware static scheduler. Finally, code generation is performed from the energy-efficient SAG to produce a multi-process program that implements the original CDFD specification. Experimental results establish the efficacy of the proposed approach.

View PDFchevron_right
Energy-aware computing for android platforms
Dr. Abhishek Agrawal

2011 International Conference on Energy Aware Computing, 2011

Android* smartphones and tablets are changing people's daily lives. People can now perform tasks on the go that were previously impossible without a laptop or desktop system. The increasingly demanding applications combined with the mobile nature of such systems place heavy emphasis on battery life. Unfortunately, many end users are not satisfied with the battery life of their Android devices. One key challenge developer's face is to understand how their software impacts energy usage. Ideally, a resource within a device should only consume power when needed and otherwise remain inactive. In this survey, we study the aggressive power management techniques underpinning the Android operating system. To aid developer's understanding of application power consumption, we introduce system-wide, function-level power profiling tools and a methodology for energy-efficiency debugging. We demonstrate the application of these analysis tools on the Android platform for common usage scenarios on a shipping Motorola Atrix* smartphone and a production Motorola XOOM* tablet.

View PDFchevron_right

References (94)

  1. Y. Agarwal, S. Hodges, R. Chandra, J. Scott, P. Bahl, and R. Gupta. Somniloquy: Augmenting Network Inter- faces to Reduce PC Energy Usage. In Proc. USENIX Symp. Networked Systems Design and Implementation (NSDI), 2009.
  2. N. Asmussen, M. Völp, B. Nöthen, H. Härtig, and G. P. Fettweis. M3: A Hardware/Operating- System Co-Design to Tame Heterogeneous Many- cores. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2016.
  3. S. Bansal and A. Aiken. Binary translation us- ing peephole superoptimizers. In Proc. USENIX Conf. Operating Systems Design and Implementation (OSDI), 2008.
  4. A. Barbalace, R. Lyerly, C. Jelesnianski, A. Carno, H.-R. Chuang, V. Legout, and B. Ravindran. Break- ing the boundaries in heterogeneous-ISA datacen- ters. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2017.
  5. A. Barbalace, M. Sadini, S. Ansary, C. Jelesnianski, A. Ravichandran, C. Kendir, A. Murray, and B. Ravin- dran. Popcorn: Bridging the Programmability Gap in heterogeneous-ISA Platforms. In Proc. The European Conf. Computer Systems (EuroSys), 2015.
  6. A. Baumann, P. Barham, P.-E. Dagand, T. Harris, R. Isaacs, S. Peter, T. Roscoe, A. Schüpbach, and A. Singhania. The Multikernel: a new OS architecture for scalable multicore systems. In Proc. ACM Symp. Operating Systems Principles (SOSP), 2009.
  7. F. Bellard. QEMU, a Fast and Portable Dynamic Trans- lator. In Proc. USENIX Annual Technical Conference (ATC), 2005.
  8. E. Blem, J. Menon, T. Vijayaraghavan, and K. Sankar- alingam. ISA wars: Understanding the relevance of ISA being RISC or CISC to performance, power, and energy on modern architectures. ACM Transactions on Computer Systems (TOCS), 33(1):3, 2015.
  9. D. Boggs, G. Brown, N. Tuck, and K. S. Venkatraman. Denver: Nvidia's First 64-bit ARM Processor. IEEE Micro, 35(2):46-55, 2015.
  10. S. Boyd-Wickizer and N. Zeldovich. Tolerating Ma- licious Device Drivers in Linux. In Proc. USENIX Annual Technical Conference (ATC), 2010.
  11. A. L. Brown and R. J. Wysocki. Suspend-to-RAM in Linux. In Ottawa Linux Symposium, 2008.
  12. X. Chen, N. Ding, A. Jindal, Y. C. Hu, M. Gupta, and R. Vannithamby. Smartphone Energy Drain in the Wild: Analysis and Implications. In Proc. ACM SIGMETRICS (SIGMETRICS), 2015.
  13. X. Chen, A. Jindal, N. Ding, Y. C. Hu, M. Gupta, and R. Vannithamby. Smartphone Background Activ- ities in the Wild: Origin, Energy Drain, and Optimiza- tion. In Proc. Ann. Int. Conf. Mobile Computing & Networking (MobiCom), 2015.
  14. B.-G. Chun, S. Ihm, P. Maniatis, M. Naik, and A. Patti. CloneCloud: Elastic Execution Between Mobile Device and Cloud. In Proc. The European Conf. Computer Systems (EuroSys), 2011.
  15. E. Cuervo, A. Balasubramanian, D.-k. Cho, A. Wol- man, S. Saroiu, R. Chandra, and P. Bahl. MAUI: mak- ing smartphones last longer with code offload. In Proc. ACM Int. Conf. Mobile Systems, Applications, & Services (MobiSys), 2010.
  16. A. d'Antras, C. Gorgovan, J. Garside, J. Goodacre, and M. Luján. HyperMAMBO-X64: Using Virtual- ization to Support High-Performance Transparent Bi- nary Translation. In Proc. Int. Conf. Virtual Execution Environments (VEE), 2017.
  17. A. d'Antras, C. Gorgovan, J. Garside, and M. Luján. Low Overhead Dynamic Binary Translation on ARM. In Proc. ACM SIGPLAN Conf. on Programming Language Design and Implementation (PLDI), 2017.
  18. M. DeVuyst, A. Venkat, and D. M. Tullsen. Execu- tion migration in a heterogeneous-ISA chip multiproces- sor. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2012.
  19. eLinux.org. PandaBoard Power Measurements. http: //elinux.org/PandaBoard_Power_Measurements.
  20. D. R. Engler, M. F. Kaashoek, and J. O'Toole, Jr. Exoker- nel: An Operating System Architecture for Application- level Resource Management. In Proc. ACM Symp. Operating Systems Principles (SOSP), 1995.
  21. P. Feiner, A. D. Brown, and A. Goel. Comprehensive ker- nel instrumentation via dynamic binary translation. In ACM SIGARCH Computer Architecture News, 2012.
  22. V. Ganapathy, M. J. Renzelmann, A. Balakrishnan, M. M. Swift, and S. Jha. The Design and Imple- mentation of Microdrivers. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2008.
  23. B. Gerofi, A. Santogidis, D. Martinet, and Y. Ishikawa. PicoDriver: Fast-path Device Drivers for Multi-kernel Operating Systems. In Proc. Int. Symp. on
  24. High-Performance Parallel and Distributed Computing (HPDC), 2018.
  25. P. Greenhalgh. Big.LITTLE processing with ARM Cortex-A15 and Cortex-A7. Technical report, 2011.
  26. M. Hähnel and H. Härtig. Heterogeneity by the numbers: A study of the ODROID XU+E big.little platform. In Y. Agarwal and K. Rajamani, editors, Proc. Workshp. Power-Aware Computing and Systems (HotPower), 2014.
  27. U. Hansson. SDIO power on/off time impacts system suspend/resume time! http://connect.linaro.org/ resource/sfo17/sfo17-402/, 2017.
  28. B. Hawkins, B. Demsky, D. Bruening, and Q. Zhao. Op- timizing Binary Translation of Dynamically Generated Code. In Proc. Int. Symp. on Code Generation and Optimization (CGO), 2015.
  29. D. Hong, C. Hsu, P. Yew, J. Wu, W. Hsu, P. Liu, C. Wang, and Y. Chung. HQEMU: a multi-threaded and retar- getable dynamic binary translator on multicores. In Proc. Int. Symp. on Code Generation and Optimization (CGO), 2012.
  30. R. J. Hookway and M. A. Herdeg. Digital FX! 32: Combining emulation and binary translation. Digital Technical Journal, 9:3-12, 1997.
  31. J. Howell, B. Parno, and J. R. Douceur. How to Run POSIX Apps in a Minimal Picoprocess. In Proc. USENIX Annual Technical Conference (ATC), 2013.
  32. Intel. Intel SuspendResume Project. https://01.org/ suspendresume, 2015.
  33. A. Kadav and M. M. Swift. Understanding Modern De- vice Drivers. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2012.
  34. A. Kantee and J. Cormack. Rump Kernels No OS? No Problem! Login: USENIX Magazine, 39(5), 2014.
  35. P. Kedia and S. Bansal. Fast Dynamic Binary Trans- lation for the Kernel. In Proc. ACM Symp. Operating Systems Principles (SOSP), 2013.
  36. A. Klaiber. The technology behind Crusoe processors. Transmeta Technical Brief, 2000.
  37. G. Kroah-Hartman. The Linux Kernel Driver Interface -Stable API Nonsense. https://www.kernel.org/doc/Documentation/ process/stable-api-nonsense.rst. (Accessed on 05/04/2019).
  38. M. Larabel. A Stable Linux Kernel API/ABI? "The Most Insane Proposal" For Linux Develop- ment. https://www.phoronix.com/scan.php?page= news_item&px=Linux-Kernel-Stable-API-ABI, 2016.
  39. M. Lentz, J. Litton, and B. Bhattacharjee. Drowsy Power Management. In Proc. ACM Symp. Operating Systems Principles (SOSP), 2015.
  40. J. LeVasseur, V. Uhlig, J. Stoess, and S. Götz. Unmod- ified Device Driver Reuse and Improved System De- pendability via Virtual Machines. In Proc. USENIX Conf. Operating Systems Design and Implementation (OSDI), 2004.
  41. T. Li, P. Brett, R. Knauerhase, D. Koufaty, D. Reddy, and S. Hahn. Operating system support for overlapping- ISA heterogeneous multi-core architectures. In Proc. IEEE Int. Symp. on High Performance Computer Architecture (HPCA), 2010.
  42. Y. Li, B. Dolan-Gavitt, S. Weber, and J. Cappos. Lock- in-Pop: securing privileged operating system kernels by keeping on the beaten path. In Proc. USENIX Annual Technical Conference (ATC), 2017.
  43. F. X. Lin, Z. Wang, R. LiKamWa, and L. Zhong. Reflex: using low-power processors in smartphones without knowing them. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2012.
  44. F. X. Lin, Z. Wang, and L. Zhong. K2: A mo- bile operating system for heterogeneous coherence do- mains. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2014.
  45. R. Liu and F. X. Lin. Understanding the Characteristics of Android Wear OS. In Proc. ACM Int. Conf. Mobile Systems, Applications, & Services (MobiSys), 2016.
  46. X. Liu, T. Chen, F. Qian, Z. Guo, F. X. Lin, X. Wang, and K. Chen. Characterizing Smartwatch Usage in the Wild. In Proceedings of the 15th Annual International Conference on Mobile Systems, Applications, and Services, 2017.
  47. LKML. [GIT PULL] PM updates for 2.6.33, 2009.
  48. D. Loghin, B. M. Tudor, H. Zhang, B. C. Ooi, and Y. M. Teo. A Performance Study of Big Data on Small Nodes. Proc. VLDB Endow., 8(7):762-773, 2015.
  49. G. Lu, J. Zhan, X. Lin, C. Tan, and L. Wang. On Hori- zontal Decomposition of the Operating System. CoRR, abs/1604.01378, 2016.
  50. C.-K. Luk, R. Cohn, R. Muth, H. Patil, A. Klauser, G. Lowney, S. Wallace, V. J. Reddi, and K. Hazel- wood. Pin: Building customized program analysis tools with dynamic instrumentation. In Proc. ACM SIGPLAN Conf. on Programming Language Design and Implementation (PLDI), 2005.
  51. LWN. Redesigning asynchronous suspend/resume. https://lwn.net/Articles/366915/, 2009.
  52. A. Madhavapeddy, R. Mortier, C. Rotsos, D. Scott, B. Singh, T. Gazagnaire, S. Smith, S. Hand, and J. Crowcroft. Unikernels: Library operating systems for the cloud. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2013.
  53. MediaTek. Microsoft Azure Sphere MCU with ex- tensive I/O peripheral subsystem for diverse IoT ap- plications. https://www.mediatek.com/products/ azureSphere/mt3620, 2018.
  54. D. Meisner, B. T. Gold, and T. F. Wenisch. PowerNap: Eliminating Server Idle Power. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2009.
  55. D. Meisner and T. F. Wenisch. DreamWeaver: architec- tural support for deep sleep. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2012.
  56. Micron Technology, Inc. TN4201 LPDDR2 Sys- tem Power Calculator. https://www.micron.com/ support/tools-and-utilities/power-calc, 2013.
  57. Mike Turquette. The Common Clk Framework. https: //www.kernel.org/doc/Documentation/clk.txt.
  58. C. Min, W. Kang, M. Kumar, S. Kashyap, S. Maass, H. Jo, and T. Kim. Solros: a data-centric operating sys- tem architecture for heterogeneous computing. In Proc. The European Conf. Computer Systems (EuroSys), 2018.
  59. J. Mogul, J. Mudigonda, N. Binkert, P. Ranganathan, and V. Talwar. Using Asymmetric Single-ISA CMPs to Save Energy on Operating Systems. IEEE Micro, 28(3):26-41, 2008.
  60. J. Morrison, D. Yang, and C. Davis. Apple watch: teardown. https://www.techinsights.com/about- techinsights/overview/blog/apple-watch- teardown/. (Accessed on 01/10/2019).
  61. N. Nethercote and J. Seward. Valgrind: A Frame- work for Heavyweight Dynamic Binary Instrumenta- tion. In Proc. ACM SIGPLAN Conf. on Programming Language Design and Implementation (PLDI), 2007.
  62. E. B. Nightingale, O. Hodson, R. McIlroy, C. Hawblitzel, and G. Hunt. Helios: heterogeneous multiprocessing with satellite kernels. In Proc. ACM Symp. Operating Systems Principles (SOSP), 2009.
  63. NXP Semiconductors. i.MX 6SoloX -fact sheet. https://www.nxp.com/docs/en/fact-sheet/ IMX6SOLOXFS.pdf. (Accessed on 05/14/2019).
  64. NXP Semiconductors. i.MX 7DS power consump- tion measurement. https://www.nxp.com/docs/en/ application-note/AN5383.pdf, 2016.
  65. NXP Semiconductors. i.MX 7 Series Applications Processors | Arm® Cortex®-A7, Cortex-M4 | NXP. https://www.nxp.com/products/processors-and- microcontrollers/arm-based-processors-and- mcus/i.mx-applications-processors/i.mx-7- processors:IMX7-SERIES, 2017. (Accessed on 05/14/2019).
  66. H. Oi. A Case Study of Energy Efficiency on a Het- erogeneous Multi-Processor. SIGMETRICS Perform. Eval. Rev., 45(2):70-72, 2017.
  67. Y. Padioleau, J. L. Lawall, R. R. Hansen, and G. Muller. Documenting and automating collateral evolutions in Linux device drivers. In J. S. Sventek and S. Hand, editors, Proc. The European Conf. Computer Systems (EuroSys), 2008.
  68. Y. Padioleau, J. L. Lawall, and G. Muller. Understanding collateral evolution in Linux device drivers. In ACM SIGOPS Operating Systems Review, 2006.
  69. N. Peters, S. Park, S. Chakraborty, B. Meurer, H. Payer, and D. Clifford. Web browser workload character- ization for power management on HMP platforms. In Proc. IEEE/ACM/IFIP International Conference on Hardware/Software Codesign and System Synthesis (CODES), 2016.
  70. A. Ponomarenko. ABI Compliance Checker. https:// lvc.github.io/abi-compliance-checker/, 2018.
  71. D. E. Porter, S. Boyd-Wickizer, J. Howell, R. Olinsky, and G. C. Hunt. Rethinking the Library OS from the Top Down. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2011.
  72. A. Reid. Trustworthy Specifications of ARM v8-A and v8-M System Level Architecture. In Proc. Formal Methods in Computer-Aided Design (FMCAD), 2016.
  73. S. Rokicki, E. Rohou, and S. Derrien. Hardware- accelerated dynamic binary translation. In Proc.
  74. ACM/IEEE Design Automation & Test in Europe Conf. (DATE), 2017.
  75. S. Rokicki, E. Rohou, and S. Derrien. Supporting runtime reconfigurable VLIWs cores through dynamic binary translation. In 2018 Design, Automation & Test in Europe Conference & Exhibition, DATE 2018, Dresden, Germany, March 19-23, 2018, 2018.
  76. Y. Shan, Y. Huang, Y. Chen, and Y. Zhang. LegoOS: A Disseminated, Distributed OS for Hardware Resource Disaggregation. In Proc. USENIX Conf. Operating Systems Design and Implementation (OSDI), 2018.
  77. H. Shen, A. Balasubramanian, A. LaMarca, and D. Wetherall. Enhancing Mobile Apps to Use Sensor Hubs Without Programmer Effort. In Proc. Int. Conf. Ubiquitous Computing (UbiComp), 2015.
  78. M. Silberstein, B. Ford, I. Keidar, and E. Witchel. GPUfs: Integrating a File System with GPUs. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2013.
  79. J. Sorber, N. Banerjee, M. D. Corner, and S. Rollins. Turducken: hierarchical power management for mobile devices. In Proc. ACM Int. Conf. Mobile Systems, Applications, & Services (MobiSys), 2005.
  80. M. M. Swift, M. Annamalai, B. N. Bershad, and H. M. Levy. Recovering Device Drivers. In Proc. USENIX Conf. Operating Systems Design and Implementation (OSDI), 2004.
  81. M. M. Swift, B. N. Bershad, and H. M. Levy. Improv- ing the Reliability of Commodity Operating Systems. In Proc. ACM Symp. Operating Systems Principles (SOSP), 2003.
  82. Texas Instruments. AM5728 Sitara Processor: Dual Arm Cortex-A15 & Dual DSP, Multimedia | TI.com. http://www.ti.com/product/AM5728. (Accessed on 05/14/2019).
  83. Texas Instruments. Cortex-M3: Proces- sor technical reference manual. http: //infocenter.arm.com/help/index.jsp?topic= /com.arm.doc.ddi0337h/index.html. (Accessed on 05/07/2019).
  84. Texas Instruments. OMAP4 Applications Processor: Technical Reference Manual. http://www.ti.com/ lit/ug/swpu235ab/swpu235ab.pdf, 2010. (Ac- cessed on 05/14/2019).
  85. D. Vasisht, Z. Kapetanovic, J. Won, X. Jin, R. Chan- dra, S. Sinha, A. Kapoor, M. Sudarshan, and S. Strat- man. FarmBeats: An IoT Platform for Data-Driven Agri- culture. In Proc. USENIX Symp. Networked Systems Design and Implementation (NSDI), 2017.
  86. VMWARE. Virtual Machine to Physical Machine Migration. https://www.vmware.com/support/v2p/ doc/V2P_TechNote.pdf, 2004.
  87. W. Wang, S. McCamant, A. Zhai, and P.-C. Yew. Enhancing Cross-ISA DBT Through Automatically Learned Translation Rules. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2018.
  88. W. Wang, P.-C. Yew, A. Zhai, S. McCamant, Y. Wu, and J. Bobba. Enabling Cross-ISA Offloading for COTS Binaries. In Proc. ACM Int. Conf. Mobile Systems, Applications, & Services (MobiSys), 2017.
  89. D. Wentzlaff and A. Agarwal. Factored operating sys- tems (fos): the case for a scalable operating system for multicores. SIGOPS Oper. Syst. Rev., 43(2):76-85, 2009.
  90. S. L. Xi, M. Guevara, J. Nelson, P. Pensabene, and B. C. Lee. Understanding the Critical Path in Power State Transition Latencies. In Proc. ACM/IEEE Int. Symp. Low Power Electronics & Design (ISLPED), 2013.
  91. C. Xu, F. X. Lin, Y. Wang, and L. Zhong. Au- tomated OS-level Device Power Management for SoCs. In Proc. ACM Int. Conf. Architectural Support for Programming Languages & Operating Systems (ASPLOS), 2015.
  92. F. Xu, Y. Liu, T. Moscibroda, R. Chandra, L. Jin, Y. Zhang, and Q. Li. Optimizing Background Email Sync on Smartphones. In Proc. ACM Int. Conf. Mobile Systems, Applications, & Services (MobiSys), 2013.
  93. S. Zhai, L. Guo, X. Li, and F. X. Lin. Decelerating Sus- pend and Resume in Operating Systems. In Proc. ACM Workshp. Mobile Computing Systems & Applications (HotMobile), 2017.
  94. Q. Zhu, M. Zhu, B. Wu, X. Shen, K. Shen, and Z. Wang. Software Engagement with Sleeping CPUs. In Proc. Workshp. Hot Topics in Operating Systems (HotOS), 2015.

Related papers

Application Transformations for Energy and Performance-Aware Device Management
Eduardo Pinheiro

2002

Energy conservation without performance degradation is an important goal for battery-operated computers, such as laptops and hand-held assistants. In this paper we determine the potential benefits of application-supported device management for optimizing energy and performance. In particular, we consider application transformations that increase device idle times and inform the operating system about the length of each upcoming period of idleness. We assess the potential energy and performance benefits of this type of application support for a laptop disk. Furthermore, we propose and evaluate a compiler framework for performing the transformations automatically for a disk device. Our experimental results demonstrate that unless applications are transformed, they cannot accrue any of the predicted benefits. In addition, they show that our compiler can produce almost the same performance and energy results that we obtain by handmodifying applications. Overall, we find that the transformations we propose can reduce disk energy consumption from 55% to 89% with only a small degradation in performance.

View PDFchevron_right
Energy characterization of embedded real-time operating systems
Andrea Acquaviva

2001

In this paper we propose a methodology to analyze the energy overhead due to the presence of an embedded operating system in a wearable device. Our objective is to determine the key parameters affecting the energy consumption of the RTOS allowing the development of more efficient OS-based power management policies. To achieve this target, we propose a characterization strategy that stimulates the RTOS both at the kernel and at the I/O driver level by analyzing various OS-related parameters. Our analisys focus in particular on the relationship between energy consumption and processor frequency characterizing the different functionalities of an RTOS, suggesting a way to develop effective OS-aware energy optimization policies based on variable voltage and frequency. Experimental results are presented for eCos, an open-source embedded OS ported and installed on a prototype of wearable device, the HP SmartBadgeIII.

View PDFchevron_right
Reducing Processor Power Consumption by Improving Processor Time Management in a Single-User Operating System
Alan Smith

1996

The CPU is one of the major power consumers in a portable computer, and considerable power can be saved by turning off the CPU when it is not doing useful work. In Apple's MacOS, however, idle time is often converted to busy waiting, and generally it is very hard to tell when no useful computation is occurring. In this paper, we suggest several heuristic techniques for identifying this condition, and for temporarily putting the CPU in a low-power state. These techniques include turning off the processor when all processes are blocked, turning off the processor when processes appear to be busy waiting, and extending real time process sleep periods. We use trace-driven simulation, using processor run interval traces, to evaluate the potential energy savings and performance impact. We find that these techniques save considerable amounts of processor energy (as much as 66%), while having very little performance impact (less than 2% increase in run time). Implementing the proposed strategies should increase battery lifetime by approximately 20% relative to Apple's current CPU power management strategy, since the CPU and associated logic are responsible for about 32% of power use; similar techniques should be applicable to operating systems with similar behavior.

View PDFchevron_right
Code Transformations for Energy-Efficient Device Management
Eduardo Pinheiro

IEEE Transactions on Computers, 2004

Energy conservation without performance degradation is an important goal for batteryoperated computers, such as laptops and hand-held assistants. In this paper we study application-supported device management for optimizing energy and performance. In particular, we consider application transformations that increase device idle times and inform the operating system about the length of each upcoming period of idleness. We use modeling and experimentation to assess the potential energy and performance benefits of this type of application support for a laptop disk. Furthermore, we propose and evaluate a compiler framework for performing the transformations automatically. Our main modeling results show that the transformations are potentially beneficial. However, our experimental results with six real laptop applications demonstrate that unless applications are transformed, they cannot accrue any of the predicted benefits. In addition, they show that our compiler can produce almost the same performance and energy results as hand-modifying applications. Overall, we find that the transformations can reduce disk energy consumption from 55% to 89% with a degradation in performance of at most 8%.

View PDFchevron_right
Power Struggles: Revisiting the RISC vs. CISC Debate on Contemporary ARM and x86 Architectures
bashier abu hassan

RISC vs. CISC wars raged in the 1980s when chip area and processor design complexity were the primary constraints and desktops and servers exclusively dominated the computing landscape. Today, energy and power are the primary design constraints and the computing landscape is significantly different: growth in tablets and smartphones running ARM (a RISC ISA) is surpassing that of desktops and laptops running x86 (a CISC ISA). Further, the traditionally low-power ARM ISA is entering the high-performance server market, while the traditionally high-performance x86 ISA is entering the mobile low-power device market. Thus, the question of whether ISA plays an intrinsic role in performance or energy efficiency is becoming important, and we seek to answer this question through a detailed measurement based study on real hardware running real applications. We analyze measurements on the ARM Cortex-A8 and Cortex-A9 and Intel Atom and Sandybridge i7 microprocessors over workloads spanning mobile, desktop, and server computing. Our methodical investigation demonstrates the role of ISA in modern microprocessors' performance and energy efficiency. We find that ARM and x86 processors are simply engineering design points optimized for different levels of performance, and there is nothing fundamentally more energy efficient in one ISA class or the other. The ISA being RISC or CISC seems irrelevant.

View PDFchevron_right
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved