CHAP Demonstration

Credit-based Home Access Point (CHAP)

by Choong-Soo Lee, Mark Claypool and Robert Kinicki

Increasing availability of high-speed Internet and wireless access points has allowed home users to connect not only their computers but various other devices to the Internet. Every device running different applications requires unique Quality of Service (QoS). It has been shown that delay-sensitive applications, such as VoIP, remote login and online game sessions, suffer increased latency in the presence of throughput-sensitive applications such as FTP and P2P. Currently, there is no mechanism at the wireless AP to mitigate these effects except explicitly classifying the traffic based on port numbers or host IP addresses. We propose CHAP, a credit-based queue management technique, to eliminate the explicit configuration process and dynamically adjust the priority of all the flows from different devices to match their QoS requirements and wireless conditions to improve application quality in home networks. An analytical model is used to analyze the interaction between flows and credits and resulting queueing delays for packets. CHAP is evaluated using Network Simulator (NS2) under a wide range of conditions against First-In-First-Out (FIFO) and Strict Priority Queue (SPQ) scheduling algorithms. CHAP improves the quality of an online game, a VoIP session, a video streaming session, and a Web browsing activity by 20%, 3%, 93%, and 51%, respectively, compared to FIFO in the presence of an FTP download. CHAP provides these improvements similar to SPQ without an explicit classification of flows and a pre-configured scheduling policy. A Linux implementation of CHAP is used to evaluate its performance in a real residential network against FIFO. CHAP reduces the web response time by up to 85% compared to FIFO in the presence of a bulk file download. Our contributions include an analytic model for the credit-based queue management, simulation, and implementation of CHAP, which provides QoS with minimal configuration at the AP.

Publications

Choong-Soo Lee, Mark Claypool, and Robert Kinicki. Implementation and Demonstration of Credit-based Home Access Point (CHAP), in Proceedings of ACM Multimedia (MM) 2010, Firenze, Italy, October 2010.
Choong-Soo Lee, Mark Claypool, and Robert Kinicki. A Credit-based Home Access Point (CHAP) to Improve Application Performance on IEEE 802.11 Networks, In Proceedings of the Multimedia Systems (MMSys), Scottsdale, Arizona, February 2010.
Choong-Soo Lee. CHAP - Credit-based Home Access Point for Overall Application QoS Improvement, the Passive and Active Measurement Conference, Seoul, South Korea, April 2009.

Posters

Choong-Soo Lee. A Credit-based Home Access Point (CHAP) to Improve Application Performance on IEEE 802.11 Networks, Graduate Research Achievement Day 2010, Worcester, MA, March, 2010.
Choong-Soo Lee. CHAP - Credit-based Home Access Point for Overall Application QoS Improvement, the Passive and Active Measurement Conference, Seoul, South Korea, April 2009.

Thesis

Choong-Soo Lee. A Credit-based Home Access Point (CHAP) to Improve Application Quality on IEEE 802.11 Networks, PhD thesis, Computer Science Department at Worcester Polytechnic Institute, 2010.

Case Study 1: Web Browsing

Please click HERE to start the videos at the same time to compare the page loading times between DropTail and CHAP

FIFO

CHAP

The videos above show that CNN website loads in about 20s and 6s with FIFO and CHAP queue respectively while downloading a DVD ISO at the same time.

The graph below shows the average webpage loading times of each website tested. FIFO queue increases the webpage loading times by a factor of 4-8 while performing a bulk file download at the same time. CHAP queue provides the same webpage loading times as FIFO queue in the absense of a bulk file download and is able to keep the webpage load times about the same even in the presence of a bulk file download.

Case Study 2a: Quake IV Server Pings

The map above depicts the locations of all the Quake IV servers available through the master server in March 2010. We used qstat to ping all the servers to record game server ping times. The ping times are shown in the graph below as CDFs.

Case Study 2b: Quake IV - Shooting a Shotgun

Quake IV Alone

Description

We test shooting a shotgun multiple times in Quake IV. This case study is designed to demonstrate how the responsiveness of Quake IV changes with respect to game server ping time.

Quake IV Alone (left) provides the baseline case with server ping times of 57-92ms. Bullets hit the wall as soon as the shotgun is fired. In the Quake IV + ISO Download (FIFO) case (bottom left), server ping times increase to 480-522ms with FIFO queue while downloading a DVD ISO. The bullets no longer reach the wall as soon as the shotgun is fired. The action and effect are not synchronized any more. In Quake IV + ISO Download (CHAP) case (bottom), server ping times stay around the same as the Quake IV Alone case at 59-115ms. The bullets do hit the wall as soon as the weapon is fired, just like in the Quake IV Alone case.

Quake IV + ISO Download (FIFO)

Quake IV + ISO Download (CHAP)

Case Study 2c: Quake IV - Moving Around

Quake IV Alone

Description

We test moving around in Quake IV. This case study is designed to demonstrate how Quake IV tries to adjust with respect to game server ping time. The objective is to move forward, pick up the weapon, and return to the original position.

Quake IV Alone (left) provides the baseline case with server ping times of 62-125ms. The game responds as expected to actions of moving forward, picking up the weapon and returning to the original position. In the Quake IV + ISO Download (FIFO) case (bottom left), server ping times increase to 477-546ms with FIFO queue while downloading a DVD ISO. Quake IV does not allow the player to move forward at all. We speculate that Quake IV times out waiting for responses to move commands from the server and keeps reseting the player back to where he was. In Quake IV + ISO Download (CHAP) case (bottom), server ping times stay around the same as the Quake IV Alone case at 62-94ms. The player successfully achieves his objective of moving forward, picking up the gun and returning to the original position.