Dark Mode
Mobile Edge Computing Task Latency Data
Data Science and Analytics
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
The collection tracks the turnaround execution times, measured in seconds, for image recognition tasks that have been offloaded to four distinct edge servers. The process begins when the edge node establishes a connection and starts sending the image, concluding when the recognition result is received back from the server. The four edge servers involved in this experiment included a MacBook Pro (1.4 GHz Quad-Core Intel Core i5), a second MacBook Pro (2.5 GHz Dual-Core Intel Core i5), an Ubuntu Virtual Machine using VirtualBox, and a Raspberry Pi 4B, each featuring unique processor and RAM configurations. The source client, a mobile edge node, was simulated as a process running on one of these devices. This resource supports research into time-optimised decision-making for task offloading.
Columns
- Time: Records the specific timestamp of the task execution, including the day, date, hours, minutes, second, and year. This column contains 1,000 unique, valid values.
- Execution Time (Turnaround Task Execution time): Represents the total duration, measured in seconds, from the beginning of the image submission until the result is returned by the edge server. Values range from 0.09 to 0.23 seconds, with a mean execution time of approximately 0.11 seconds across the records.
Distribution
The file is supplied in a flat data file format (MacBookPro1.csv), with a size of 44.84 kB. It comprises 2 columns and contains 1,000 recorded observations or records. All data points in both fields are validated, with no missing or mismatched values.
Usage
This data is ideally suited for:
- Developing and testing algorithms for time-optimised task offloading in mobile edge computing environments.
- Benchmarking the performance and latency characteristics of different hardware platforms (MacBook Pros, Ubuntu VMs, Raspberry Pi 4B) when executing computer vision tasks.
- Evaluating resource allocation and scheduling strategies in distributed systems.
- Researching real-time decision-making processes for resource-constrained edge devices.
Coverage
The data scope is based on temporal recordings of execution times, captured during laboratory or simulated environments involving specific, named server hardware configurations. It focuses purely on technical performance metrics and has no geographical or demographic scope. The associated introductory paper, Time-Optimized Task Offloading Decision Making in Mobile Edge Computing, was published in 2019. This dataset is static, with no expected updates.
License
Attribution 4.0 International (CC BY 4.0)
Who Can Use It
- Computer Science Researchers: For developing models to reduce latency in distributed systems.
- Engineers specialising in Mobile and Wireless Systems: To understand the practical impact of hardware heterogeneity on MEC task performance.
- Academics focused on Computer Vision: To analyse task execution speed across varied processing platforms.
- Students: As a foundation for projects involving performance analysis of edge computing infrastructure.
Dataset Name Suggestions
- Edge Server Task Execution Times
- Mobile Edge Computing Task Latency Data
- Heterogeneous Edge Server Performance Log
- Image Recognition Offload Benchmarks
Attributes
Original Data Source: Mobile Edge Computing Task Latency Data
QUALITYLoading...