![On Availability-Performability Tradeoff in Wireless Mesh Networks
Abstract:
It is understood from past decade of research that a wireless multi-hop
network can achieve maximum network throughput only when its nodes
operate at a minimum common transmission power level that ensures
network connectivity (availability). This point of optimality where
maximum availability and throughput is guaranteed in an interference-
optimal network has been the basis of numerous design problems in
wireless networks. In this paper, we claim that when performability
(availability weighted performance) is considered as opposed to average
case throughput performance, there does not exist a transmission power
(or node density) that can maximize both availability and performability.
Since the current mesh networks are expected to deliver carrier-grade
services to its users, the availability-performability tradeoff presented in
this paper holds a special importance. While availability metric is a
necessary one for any networking system intended to provide continuous
service, past research [1] has shown a strong correlation between
performability and quality of user experience in case of wireless networks.
The contributions of the paper are as follows: (1) We first define availability
and performability in the context of wireless mesh networks, and then
develop efficient algorithms on the basis of intelligent state sampling that](https://image.slidesharecdn.com/onavailability-performabilitytradeoffinwirelessmeshnetworks-150623051102-lva1-app6891/85/On-availability-performability-tradeoff-in-wireless-mesh-networks-1-320.jpg)
![can calculate both the quantities with reasonable accuracy. (2) We apply
the evaluation methods to two existing mesh networks (GoogleWiFi [2]
and PoncaCityMesh [3]) to demonstrate that their current design can not
guarantee a reasonable level of availability or performability. (3) Using
hundreds of hours of simulations, we analyze the impact of two basic
deployment factors (node density and transmission power) on availability
and performability. We outline numerous novel results that emerge due to
joint availability-performability analysis including the observation about
availability-performability tradeoff.
Existing System:
The performability and availability analysis holds a special importance for
WMNs because current mesh deployments are expected to deliver carrier-
grade services to its users. Current WMNs are being designed to handle
emergency services, surveillance, smart-grid operations etc. It is absolutely
essential that a mesh network continues to operate with high availability
and performability even in presence of random link or node failures.
With these increasing expectations, it is generally being recognized that
service continuity metrics such as availability and performability are not
well-understood for WMNs. This is natural since ad-hoc networks (from
which the mesh concept was derived) were never designed to provide
carrier-grade services. WMNs, on the other hand, must be analyzed in the
same context as other contenders such as cellular networks.
Proposed System:
We first define availability and performability in the context of WMNs.
While their definitions are more or less intuitive, we show that evaluating
them precisely is extremely complex. We then present efficient algorithms](https://image.slidesharecdn.com/onavailability-performabilitytradeoffinwirelessmeshnetworks-150623051102-lva1-app6891/85/On-availability-performability-tradeoff-in-wireless-mesh-networks-2-320.jpg)
On availability performability tradeoff in wireless mesh networks
- 1. On Availability-Performability Tradeoff in Wireless Mesh Networks Abstract: It is understood from past decade of research that a wireless multi-hop network can achieve maximum network throughput only when its nodes operate at a minimum common transmission power level that ensures network connectivity (availability). This point of optimality where maximum availability and throughput is guaranteed in an interference- optimal network has been the basis of numerous design problems in wireless networks. In this paper, we claim that when performability (availability weighted performance) is considered as opposed to average case throughput performance, there does not exist a transmission power (or node density) that can maximize both availability and performability. Since the current mesh networks are expected to deliver carrier-grade services to its users, the availability-performability tradeoff presented in this paper holds a special importance. While availability metric is a necessary one for any networking system intended to provide continuous service, past research [1] has shown a strong correlation between performability and quality of user experience in case of wireless networks. The contributions of the paper are as follows: (1) We first define availability and performability in the context of wireless mesh networks, and then develop efficient algorithms on the basis of intelligent state sampling that
- 2. can calculate both the quantities with reasonable accuracy. (2) We apply the evaluation methods to two existing mesh networks (GoogleWiFi [2] and PoncaCityMesh [3]) to demonstrate that their current design can not guarantee a reasonable level of availability or performability. (3) Using hundreds of hours of simulations, we analyze the impact of two basic deployment factors (node density and transmission power) on availability and performability. We outline numerous novel results that emerge due to joint availability-performability analysis including the observation about availability-performability tradeoff. Existing System: The performability and availability analysis holds a special importance for WMNs because current mesh deployments are expected to deliver carrier- grade services to its users. Current WMNs are being designed to handle emergency services, surveillance, smart-grid operations etc. It is absolutely essential that a mesh network continues to operate with high availability and performability even in presence of random link or node failures. With these increasing expectations, it is generally being recognized that service continuity metrics such as availability and performability are not well-understood for WMNs. This is natural since ad-hoc networks (from which the mesh concept was derived) were never designed to provide carrier-grade services. WMNs, on the other hand, must be analyzed in the same context as other contenders such as cellular networks. Proposed System: We first define availability and performability in the context of WMNs. While their definitions are more or less intuitive, we show that evaluating them precisely is extremely complex. We then present efficient algorithms
- 3. that can estimate the quantities with bounded confidence. Both, availability and performability, are then studied for two basic topology deployment factors—transmission power and node density. We believe that this evaluation and analysis of availability-performability trade-off is crucial in understanding how a realistic service-level agreement (SLA) can be established for current-age WMNs. We target tx-power and node deployment in our study because their impact on topology formation and network capacity is the most fundamental in nature. It has been shown that other recent ways of performance improvement (such as channel assignment, MIMO etc.) remain orthogonal to their impact. Hardware Requirements: • System : Pentium IV 2.4 GHz. • Hard Disk : 40 GB. • Floppy Drive : 1.44 Mb. • Monitor : 15 VGA Colour. • Mouse : Logitech. • RAM : 256 Mb. Software Requirements: • Operating system : - Windows XP. • Front End : - JSP • Back End : - SQL Server
- 4. Software Requirements: • Operating system : - Windows XP. • Front End : - .Net • Back End : - SQL Server