26-02-2013, 04:35 PM
Latency Increases for CDN Using LatLong Tool
Abstract
Content Distribution Networks (CDNs) are increasingly being used to disseminate data in today's Internet aiming at reducing the load on the origin server and the traffic on the Internet, and ultimately improving response time to users. In this direction, crucial data management issues should be addressed. Latency may increase for many reasons, such as interdomain routing changes and the CDN’s own load-balancing policies. Here, we propose a tool for CDNs to diagnose large latency increases, based on passive measurements of performance, traffic, and routing. We make a decision tree for classifying latency changes, and determine how to distinguish traffic shifts from increases in latency for existing servers, routers, and paths. Another challenge is that network operators group related clients to reduce measurement and control overhead, but the clients in a region may use multiple servers and paths during a measurement interval. We propose metrics that quantify the latency contributions across sets of servers and routers. Based on the design, we implement the LatLong tool for diagnosing large latency increases for CDN. To detect and diagnose latency problems, CDNs could deploy a large-scale active-monitoring infrastructure to collect performance measurements from synthetic clients all over the world. We show that the proposed technique can yield up to 25% reduction in user-perceived latency, compared with other heuristic schemes which have knowledge of the content popularity.
EXISTING SYSTEM:
Most clients are served by a geographically nearby CDN node, a sizeable fraction of clients experience latencies several tens of milliseconds higher than other clients in the same region.
CDNs are quite vulnerable to increases in the wide-area latency between their servers and the clients, due to interdomain routing changes or congestion in other domains.
Finding the root cause of latency increases is difficult. Many factors can contribute to higher delays, including internal factors like how the CDN selects servers for the clients, and external factors such as interdomain routing changes. Moreover, separating cause from effect is a major challenge.
WhyHigh uses active measurements, combined with routing and traffic data, to study persistent performance problems where some clients in a geographic region have much higher latency than others.
WhyHigh helps in diagnosing the cause of several instances of path inflation, it is not perfect. This is a consequence of WhyHigh’s partial view of Internet routing.
PROPOSED SYSTEM:
We use passive measurements to analyze large latency changes affecting entire groups of clients. The dynamics of latency increases caused by changes in server selection and inter-domain routing are not studied in the work of WhyHigh.
This tool explores how CDNs can diagnose latency problems based on measurements they can readily and efficiently collect—passive measurements of performance, traffic, and routing from their own networks.
We propose a decision tree for separating the causes of latency changes from their effects, and identify the data sets needed for each step in the analysis. We analyze the measurement data to identify suitable thresholds to identify large latency changes and to distinguish one possible cause from another.
Replicating content across a geographically distributed set of servers and redirecting every client to the server with least latency is commonly expected to significantly improve client performance.
System Requirement Specifications
Hardware Requirements:
• PIV 2.8 GHz Processor and Above
• RAM 512MB and Above
• HDD 40 GB Hard Disk Space and Above
Software Requirements:
• WINDOWS OS (XP / 2000 / 200 Server / 2003 Server)
• Visual Studio .Net 2010 Enterprise Edition
• Visual Studio .Net Framework (Minimal for Deployment) version 4.0
• SQL Server 2005