Loading...

Modeling Dependent Requests in Caching Mechanisms and Its Effect on Delay and Age of Information

Rahnamania, Mehran | 2024

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58014 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Ashtiani, Farid
  7. Abstract:
  8. In many communication scenarios, dependent requests are commonly observed, such as caching networks. In this thesis, we propose a new analytical model to examine the impact of dependent arrivals on the average response time in a simple queue with exponential service time. We examine two dependency scenarios: batch arrivals and the arrival of one customer after another with a random delay with a known distribution. To accurately model the effect of dependency between arrivals, we define redundant customers in our model to compensate for the existing dependency effects. We demonstrate the validity of our modeling approach analytically in the first scenario and through simulation in the second scenario. This model can be used in various communication scenarios for system performance analysis. Additionally, we examine an example of dependent arrivals in a communication scenario. We consider a simple cloud-edge caching system for highly dynamic content, consisting of a cache in the cloud (content provider) and a cache at the edge (base station). Each file is divided into two parts. In case of a file request by any user, its components are sent to the user such that the part in the cache is push-based (regularly updated and stored in the cache) and the other part is pull-based (fetched from the cloud cache by the base station and then sent to the end user). Thus, half of the files are stored in the cache and are randomly updated, and the other half need to be fetched before being sent to the user. We derive the average delay and the average age of information for three caching and delivery schemes: whole caching whole delivery (WCWD), partial caching whole delivery (PCWD), and partial caching partial delivery (PCPD). To this end, we propose a product-form queueing network model with redundant arrivals to model the dependent arrivals resulting from simultaneous fetching and parallel retrieval of both parts of the files (due to file splitting and partial caching). Our simulations show the high accuracy of the analytical model. We will observe that PCPD is superior to the other schemes in various conditions
  9. Keywords:
  10. Cache Memory ; Queueing System ; Partial Caching ; Dependent Arrivals ; Edge-Cloud Computing

 Digital Object List

 Bookmark

No TOC