Phan, Linh T.X.

Email Address
ORCID
Disciplines
Research Projects
Organizational Units
Position
Introduction
Research Interests

Search Results

Now showing 1 - 10 of 35
  • Publication
    Video Quality Driven Buffer Sizing via Frame Drops
    (2011-08-01) Gangadharan, Deepak; Phan, Linh T.X.; Chakraborty, Samarjit; Zimmermann, Roger; Lee, Insup
    We study the impact of video frame drops in buffer constrained multiprocessor system-on-chip (MPSoC) platforms. Since on-chip buffer memory occupies a significant amount of silicon area, accurate buffer sizing has attracted a lot of research interest lately. However, all previous work studied this problem with the underlying assumption that no video frame drops can be tolerated. In reality, multimedia applications can often tolerate some frame drops without significantly deteriorating their output quality. Although system simulations can be used to perform video quality driven buffer sizing, they are time consuming. In this paper, we first demonstrate a dual-buffer management scheme to drop only the less significant frames. Based on this scheme, we then propose a formal framework to evaluate the buffer size vs. video quality trade-offs, which in turn will help a system designer to perform quality driven buffer sizing. In particular, we mathematically characterize the maximum numbers of frame drops for various buffer sizes and evaluate how they affect the worst-case PSNR value of the decoded video. We evaluate our proposed framework with an MPEG-2 decoder and compare the obtained results with that of a cycle-accurate simulator. Our evaluations show that for an acceptable quality of 30 dB, it is possible to reduce the buffer size by upto 28.6% which amounts to 25.88 megabits.
  • Publication
    On the Feasibility of Dynamic Rescheduling on the Intel Distributed Computing Platform
    (2010-11-01) Phan, Linh T.X.; Zhang, Zhuoyao; Tan, Godfrey; Jain, Saumya; Duong, Harrison; Lee, Insup; Loo, Boon Thau
    This paper examines the feasibility of dynamic rescheduling techniques for effectively utilizing compute resources within a data center. Our work is motivated by practical concerns of Intel’s NetBatch system, an Internet-scale data center based distributed computing platform developed by Intel Corporation for massively parallel chip simulations within the company. NetBatch has been operational for many years, and currently is deployed live on tens of thousands of machines that are globally distributed at various data centers. We perform an analysis of job execution traces obtained over a one year period collected from tens of thousands of NetBatch machines from 20 different pools. Our analysis show that we observe that the NetBatch currently does not make full use of all the resources. Specifically, the job completion time can be severely impacted due to job suspension when higher priority jobs preempt lower priority jobs. We then develop dynamic job rescheduling strategies that adaptively restart jobs to available resources elsewhere, which better utilize system resources and improve completion times. Our trace-driven evaluation results show that dynamic rescheduling enables NetBatch to significantly reduce system waste and completion time of suspended jobs.
  • Publication
    Mixed-Criticality Scheduling on Multiprocessors using Task Grouping
    (2015-07-01) Ren, Jiankang; Phan, Linh T.X.
    Real-time systems are increasingly running a mix of tasks with different criticality levels: for instance, unmanned aerial vehicle has multiple software functions with different safety criticality levels, but runs them on a single, shared computational platform. In addition, these systems are increasingly deployed on multiprocessor platforms because this can help to reduce their cost, space, weight, and power consumption. To assure the safety of such systems, several mixed-criticality scheduling algorithms have been developed that can provide mixed-criticality timing guarantees. However, most existing algorithms have two important limitations: they do not guarantee strong isolation among the high-criticality tasks, and they offer poor real-time performance for the low-criticality tasks.
  • Publication
    MC-ADAPT: Adaptive Task Dropping in Mixed-Criticality Scheduling
    (2017-10-01) Lee, Jaewoo; Phan, Linh T.X.; Chwa, Hoon Sung; Lee, Insup; Shin, Insik
    Recent embedded systems are becoming integrated systems with components of different criticality. To tackle this, mixed-criticality systems aim to provide different levels of timing assurance to components of different criticality levels while achieving efficient resource utilization. Many approaches have been proposed to execute more lower-criticality tasks without affecting the timeliness of higher-criticality tasks. Those previous approaches however have at least one of the two limitations; i) they penalize all lower-criticality tasks at once upon a certain situation, or ii) they make the decision how to penalize lowercriticality tasks at design time. As a consequence, they underutilize resources by imposing an excessive penalty on lowcriticality tasks. Unlike those existing studies, we present a novel framework, called MC-ADAPT, that aims to minimally penalize lower-criticality tasks by fully reflecting the dynamically changing system behavior into adaptive decision making. Towards this, we propose a new scheduling algorithm and develop its runtime schedulability analysis capable of capturing the dynamic system state. Our proposed algorithm adaptively determines which task to drop based on the runtime analysis. To determine the quality of task dropping solution, we propose the speedup factor for task dropping while the conventional use of the speedup factor only evaluates MC scheduling algorithms in terms of the worst-case schedulability. We apply the speedup factor for a newly-defined task dropping problem that evaluates task dropping solution under different runtime scheduling scenarios. We derive that MC-ADAPT has a speedup factor of 1.619 for task drop. This implies that MC-ADAPT can behave the same as the optimal scheduling algorithm with optimal task dropping strategy does under any runtime scenario if the system is sped up by a factor of 1.619.
  • Publication
    Holistic resource allocation for multicore real-time systems
    (2019-04-01) Phan, Linh T.X.; Xu, Meng; Choi, Hyon-Young; Lin, Yuhan; Li, Haoran; Lee, Insup; Lu, Chenyang
    This paper presents CaM, a holistic cache and memory bandwidth resource allocation strategy for multicore real-time systems. CaM is designed for partitioned scheduling, where tasks are mapped onto cores, and the shared cache and memory bandwidth resources are partitioned among cores to reduce resource interferences due to concurrent accesses. Based on our extension of LITMUSRT with Intel’s Cache Allocation Technology and MemGuard, we present an experimental evaluation of the relationship between the allocation of cache and memory bandwidth resources and a task’s WCET. Our resource allocation strategy exploits this relationship to map tasks onto cores, and to compute the resource allocation for each core. By grouping tasks with similar characteristics (in terms of resource demands) to the same core, it enables tasks on each core to fully utilize the assigned resources. In addition, based on the tasks’ execution time behaviors with respect to their assigned resources, we can determine a desirable allocation that maximizes schedulability under resource constraints. Extensive evaluations using real-world benchmarks show that CaM offers near optimal schedulability performance while being highly efficient, and that it substantially outperforms existing solutions.
  • Publication
    An Empirical Analysis of Scheduling Techniques for Real-Time Cloud-Based Data Processing
    (2011-12-01) Phan, Linh T.X.; Loo, Boon Thau; Zhang, Zhuoyao; Lee, Insup; Zheng, Qi
    In this paper, we explore the challenges and needs of current cloud infrastructures, to better support cloud-based data-intensive applications that are not only latency-sensitive but also require strong timing guarantees. These applications have strict deadlines (e.g., to perform time-dependent mission critical tasks or to complete real-time control decisions using a human-in-the-loop), and deadline misses are undesirable. To highlight the challenges in this space, we provide a case study of the online scheduling of MapReduce jobs executed by Hadoop. Our evaluations on Amazon EC2 show that the existing Hadoop scheduler is ill-equipped to handle jobs with deadlines. However, by adapting existing multiprocessor scheduling techniques for the cloud environment, we observe significant performance improvements in minimizing missed deadlines and tardiness. Based on our case study, we discuss a range of challenges in this domain posed by virtualization and scale, and propose our research agenda centered around the application of advanced real-time scheduling techniques in the cloud environment.
  • Publication
    Modeling Buffers With Data Refresh Semantics in Automotive Architectures
    (2010-10-01) Phan, Linh T.X.; Schneider, Reinhard; Chakraborty, Samarjit; Lee, Insup
    Automotive architectures consist of multiple electronic control units (ECUs) which run distributed control applications. Such ECUs are connected to sensors and actuators and communicate via shared buses. Resource arbitration at the ECUs and also in the communication medium, coupled with variabilities in execution requirements of tasks results in jitter in the signal/data streams existing in the system. As a result, buffers are required at the ECUs and bus controllers. However, these buffers often implement different semantics – FIFO queuing, which is the most straightforward buffering scheme, and data refreshing, where stale data is overwritten by freshly sampled data. Traditional timing and schedulability analysis that are used to compute, e.g., end-to-end delays, in such automotive architectures can only model FIFO buffering. As a result, they return pessimistic delay and resource estimates because in reality overwritten data items do not get processed by the system. In this paper we propose an analytical framework for accurately modeling such data refresh semantics. Our model exploits a novel feedback control mechanism and is purely functional in nature. As a result, it is scalable and does not involve any explicit state modeling. Using this model we can estimate various timing and performance metrics for automotive ECU networks consisting of buffers implementing different data handling semantics. We illustrate the utility of this model through three case studies from the automotive electronics domain.
  • Publication
    Cache-aware Interfaces for Compositional Real-Time Systems
    (2015-12-01) Phan, Linh T.X.; Xu, Meng; Lee, Insup
    Interface-based compositional analysis is by now a fairly established area of research in real-time systems. However, current research has not yet fully considered practical aspects, such as the effects of cache interferences on multicore platforms. This position paper discusses the analysis challenges and motivates the need for cache scheduling in this setting, and it highlights several research questions towards cache-aware interfaces for compositional systems on multicore platforms.
  • Publication
    Platform-Dependent Code Generation for Embedded Real-Time Software
    (2013-09-29) Kim, BaekGyu; Phan, Linh T.X.; Sokolsky, Oleg; Lee, Insup
    Code generation for embedded systems is challenging, since the generated code (e.g., C code) is expected to run on a heterogeneous set of target platforms with different characteristics, such as hardware/software architectures and programming interfaces. We propose a code generation framework that provides the flexibility to generate different source code that is executable on each target platform. In our framework, the platform-dependent characteristics of a target platform are explicitly specified by an Architectural Analysis Description Language (AADL) model and a code snippet repository. The AADL model captures hardware/software architectural aspects of the platform, such as periodic/aperiodic threads and their interactions with sensors and actuators. The code snippet repository contains platform-dependent code snippets that are categorized according to the functions required to implement the components of the AADL model. These two elements of the platform capability are then used by the code generation algorithm to generate platform-dependent code for the given platform. We demonstrate the applicability of our framework using a case study of code generation for two infusion pump systems.
  • Publication
    Compositional Analysis of Real-Time Embedded Systems
    (2011-10-01) Phan, Linh T.X.; Lee, Insup; Sokolsky, Oleg
    This tutorial is concerned with various aspects of component-based design and compositional analysis of real-time embedded systems. It will first give an overview of component-based frameworks and their underlying principles. It will then go in-depth into abstraction methods for real-time components and techniques for computing their optimal interfaces, for both systems implemented on uniprocessor and multiprocessor platforms, as well as extensions to multi-mode systems. Besides theoretical aspects, the tutorial will also present an implementation of the compositional analysis framework on Xen virtualization and a demonstration of the CARTS toolset with several examples seeing the techniques in action. It will also include two case studies highlighting the utility of the framework, including the ARINC-653 avionics software and a smart-phone application. We will conclude the tutorial with a number of open challenges and research opportunities in this domain.