Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimize Replica Server Placement in a Satellite Network

Zhiyuan He, Yi Xu, Cheng Luo, Lili Qiu, Yuqing Yang

TL;DR

This work proposes placing content replica servers within satellite networks and optimizing replica placement for important performance metrics, such as latency, transmission, and storage cost, and demonstrates the effectiveness of this approach using both simulated traffic traces and a prototype system.

Abstract

Satellite communication offers Internet connectivity to remote locations, such as villages, deserts, mountains, and at sea. However, transmitting content over satellite networks is significantly more expensive than traditional Internet. To address this issue, we propose placing content replica servers within satellite networks and optimizing replica placement for important performance metrics, such as latency, transmission, and storage cost. Our approach can support different types of satellite networks, including Low Earth Orbit (LEO), Medium Earth Orbit (MEO), Geostationary Orbit (GEO), and their combinations. An important challenge for supporting content replicas in such networks is that LEO and MEO satellites are constantly moving. We address this challenge by explicitly considering their moving trajectories and strategically optimizing not only client performance, but also the cost of transferring content from one satellite to another as needed. We demonstrate the effectiveness of our approach using both simulated traffic traces and a prototype system.

Optimize Replica Server Placement in a Satellite Network

TL;DR

This work proposes placing content replica servers within satellite networks and optimizing replica placement for important performance metrics, such as latency, transmission, and storage cost, and demonstrates the effectiveness of this approach using both simulated traffic traces and a prototype system.

Abstract

Satellite communication offers Internet connectivity to remote locations, such as villages, deserts, mountains, and at sea. However, transmitting content over satellite networks is significantly more expensive than traditional Internet. To address this issue, we propose placing content replica servers within satellite networks and optimizing replica placement for important performance metrics, such as latency, transmission, and storage cost. Our approach can support different types of satellite networks, including Low Earth Orbit (LEO), Medium Earth Orbit (MEO), Geostationary Orbit (GEO), and their combinations. An important challenge for supporting content replicas in such networks is that LEO and MEO satellites are constantly moving. We address this challenge by explicitly considering their moving trajectories and strategically optimizing not only client performance, but also the cost of transferring content from one satellite to another as needed. We demonstrate the effectiveness of our approach using both simulated traffic traces and a prototype system.
Paper Structure (25 sections, 3 equations, 12 figures, 6 tables, 2 algorithms)

This paper contains 25 sections, 3 equations, 12 figures, 6 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Background
  3. Content Distribution Network (CDN)
  4. Satellite Networks
  5. Our Approach
  6. Problem Formulation
  7. Our Replication Algorithms
  8. Multi-time Local Search
  9. Multi-time Orbit-based Local Search
  10. Evaluation
  11. Experiment Setup
  12. Algorithm Evaluation
  13. Baseline methods
  14. Results of Different Metrics
  15. Oracle v.s. Prediction of Demand
  16. ...and 10 more sections

Figures (12)

  • Figure 1: A typical path between a client and the Internet in LEO satellite network.
  • Figure 2: Measured download time of web homepage and latency to CDN in satellite network and terrestrial network.
  • Figure 3: Evaluation of different methods based on 3 real traces (MAWI, Wikipedia, CAIDA) is presented in the figure. The total cost of various metrics is displayed in columns.
  • Figure 4: This figure shows the breakdown of the total cost: Query cost, replication cost and storage cost. The metric is hop count.
  • Figure 5: Evaluation of methods without oracle information of demand. Predicted demand based on average historical values is used for each method.
  • ...and 7 more figures