Exploring storage bottlenecks in Linux-based embedded systems

Abstract

With recent advances in non-volatile memory technologies and embedded hardware, large, high-speed persistent-storage devices can now realistically be used in embedded systems. Traditional models of storage systems, including the implementation in the Linux kernel, assume the performance of storage devices to be far slower than CPU and system memory speeds, encouraging extensive caching and buffering over direct access to storage hardware. In an embedded system, however, processing and memory resources are limited while storage hardware can still operate at full speed, causing this balance to shift, and leading to the observation of performance bottlenecks caused by the operating system rather than the speed of storage devices themselves. In this paper, we present performance and profiling results from high-speed storage devices attached to a Linux-based embedded system, showing that the kernel's standard file I/O operations are inadequate for such a set-up, and that `direct I/O' may be preferable for certain situations. Examination of the results identifies areas where potential improvements may be made in order to reduce CPU load and increase maximum storage throughput.

Application of Virtual Instrument Technology in the Teaching of Embedded System Course

“…Students use computer simulation software to simulate experiments. Observe, analyze, and process the simulation results through the virtual measuring instrument system [2]. Because simulation software and virtual equipment have human-computer interaction capabilities, in this virtual interactive simulation world, a kind of almost real experimental feeling is produced [17].…”

“…In this article, they gave the performance and evaluation results of high-speed storage devices connected to Linux-based embedded systems, indicating that the kernel's standard file I/O operations are not suitable for this setting, and in some cases, "direct I/O" may be preferable. However, their experiments lack certain accuracy, and more experimental data are needed to support them [2]. Oklahoma's existing nonvolatile memory (nvm) has many attractive features as the main memory of an embedded system.…”

Software Simulation in English Translation Platform Based on Gaussian Hybrid Model

“…Based on Related Technology smartphones as the representative, many smart devices have appropriate restrictions on the response speed, hardware allocation, power capacity, memory size, and other factors due to their portability and mobility requirements [19]. erefore, users will have a better experience with online translation based on the local translation in the thesaurus.…”

Embedded System Intelligent Platform Design Based on Digital Multimedia Artistic Design

“…By examining the results, you can determine where improvements can be made to reduce the CPU load and increase the maximum storage throughput. However, the performance of the research model he designed is not good enough [3].…”