Bandwidth Issues - IT Library : ZDNet Asia

White Papers

Case Studies

Vendor Webcast

Show all content

Title Date Added Company

Group-Aware Stream Filtering for Bandwidth-Efficient Data Dissemination 2008-03-06 Dartmouth College

This paper is concerned with disseminating high-volume data streams to many simultaneous applications over a low-bandwidth wireless mesh network. For bandwidth efficiency, they propose a group-aware stream filtering approach, used in conjunction with multicasting, that exploits two overlooked, yet important, properties of these applications, many applications can tolerate some degree of "Slack" in their data quality requirements, and there may exist multiple subsets of the source data satisfying the quality needs of an application. This paper also analyzes the key factors that affect its performance, based on testing with heterogeneous filtering requirements.

Tags: Data Quality

Humble ISD's Centralized, Clustered Network Infrastructure Is Improving the Quality of Education Through 20-Fold Faster Data Access, Greater Uptime, More Reliable Data Backup and a 40% Reduction in IT Administration Time 2008-03-01 Dell

The Humble Independent School District is named for a pioneering Texas settler, not for its approach to education. The fast-growing Humble Independent School District needed to improve network bandwidth and uptime, make data backup faster and more reliable, and reduce the time its IT staff spent on repairs and routine maintenance. Dell Services helped Humble ISD implement a highly robust, scalable server cluster, based on Dell PowerEdge servers. The cluster consolidates e-mail, office software and educational tools from 40 separate servers and offers dramatically faster access and simplified maintenance. The district turned its old data center into a mirrored disaster recovery site using Dell and EMC technology.

Tags: Application Servers, Back-up

Microsoft Dynamics CRM 4.0: Performance and Scalability - Bandwidth Utilization Improvements 2008-03-01 Microsoft

Microsoft Dynamics CRM business software is designed to help enterprise organizations achieve a 360-degree view of their customers across marketing, sales, and service. Engineered to deliver performance that meets the needs of the largest global deployments, Microsoft Dynamics CRM has been tested for user scalability, data scalability, and network performance. This white paper focuses on network scalability improvements.

Microsoft has made significant enhancements to the architecture of Microsoft Dynamics CRM 4.0 to improve network performance. Test results demonstrate network bandwidth utilization improvements of up to 94%. Improvements were evaluated in the number of round trips, bytes sent, and bytes received. Because the bandwidth utilization associated with a page can change after the first time the page is loaded, both initial (cold) and subsequent (warm) page loads were tested.

Tags: Scalability, CRM Software, Network Design, High Performance Computing

A Novel Technique for SNMP Bandwidth Reduction: Simulation and Evaluation 2008-02-01 Taif University

One of the most popular and widely used Network Management Protocols (NMP) is SNMP. SNMP is extremely simple and versatile but it has scalability problems. A number of alternative NMPs such as RMI interfaces, CORBA and RMON have been attempted to overcome the problems with SNMP. However, these interfaces lack the standardization effort and simplicity of the usage that SNMP provides. This paper demonstrates a new technique to decrease a number of messages between manager and agent in the SNMP (i.e. traffic reduction). The base idea of the technique is to make a system manager keeps the MIB objects that are frequently required. They simulate the proposed technique using a network simulator called NS2.

Tags: TCP - IP, Network Administration

Bandwidth Constrained IP Multicast Traffic Engineering Without MPLS Overlay 2008-01-01 University of Surrey

Existing multicast Traffic Engineering (TE) solutions tend to use explicit routing through MPLS tunnels. This paper shifts away from this overlay approach and address the bandwidth constrained IP multicast TE directly based on link state IGP routing protocols. The objective is that, through plain PIM-SM shortest path routing with optimized Multi-topology IS-IS (MISIS) link weights, the resulting multicast trees are geared towards minimal consumption of bandwidth resources. The paper applies Genetic Algorithms (GA) to the calculation of optimized M-ISIS link weights that specifically cater for engineered PIM-SM routing with bandwidth guarantees.

Tags: Multicasting, MPLS

New Algorithm for Effective Utilization of Bandwidth for Sensitive Applications 2008-01-01 Indian Institute of Science

This paper proposes an algorithm, which allocates the bandwidth in an effective manner for sensitive applications. The main idea of this work is to allocate more amount of bandwidth for particular client, who has received more votes in the network. The proposed model reduces the wastage of bandwidth and increases the throughput.

Tags: Software Engineering

ABwE: A Practical Approach to Available Bandwidth Estimation 2008-01-01 Stanford University

This paper reports on a new tool developed for monitoring available bandwidth in the range from several Mbps to 1000 Mbps. The tool is based on the packet pair dispersion technique. It has been in experimental use at SLAC for several months and it has been demonstrated at iGrid2002 and SC2002. It can be used for detecting bandwidth changes caused by routing or congestions. This paper has studied the influence of cross traffic and the behavior of routers on the delay of the probing packets. The paper presents part of these analyses together with results from monitoring of 12 paths in US, Japan and in Europe.

Tags: TCP - IP

Achieving High Throughput in Low Multiplexed, High Bandwidth, High Delay Environments 2008-01-01 Columbia University

This paper explores combining ECN bits and loss feedback to achieve a high throughput connection in a high bandwidth, low multiplexed environment. In the scheme, a TCP source reacts differently on receiving an ECN than it does on inferring a packet loss in terms of congestion window update. Coupled with an AQM scheme such as RED, TCP sources achieve high utilization at a bottleneck router, but at the cost of fairness. The paper introduces a lightweight AQM mechanism that ensures fairness while still providing high utilization to the sources. The proposal is complementary to the HSTCP modification suggested by Floyd et al.

Tags: TCP - IP

Trickle: A Userland Bandwidth Shaper for Unix-Like Systems 2008-01-01 monkey.org

As with any finite resource, it is often necessary to apply policies to the shared usage of network resources. Existing solutions typically implement this by employing traffic management in edge routers. However, users of smaller networks regularly find themselves in need of nothing more than ad-hoc rate limiting. Furthermore, the networks in question are typically unmanaged, with no network administrator(s) to manage complicated traffic management schemes. Trickle fills this gap by providing a simple and portable solution to rate limit the TCP connections of a given process or group of processes. Trickle works by taking advantage of the Unix dynamic loader's preloading functionality to interposition itself in front of the libc API interfaces that are used for sending and receiving data through BSD network sockets.

Tags: TCP - IP, UNIX

Measuring End-to-End Bandwidth With Iperf Using Web100 2008-01-01 Stanford University

End-to-end bandwidth estimation tools like Iperf though fairly accurate are intrusive. This paper describes how with an instrumented TCP stack (Web100), one can estimate the end-to-end bandwidth accurately, while consuming significantly less network bandwidth and time. The paper modified Iperf to use Web100 to detect the end of slow-start and estimate the end-to-end bandwidth by measuring the amount of data sent for a short period (1 second) after the slow-start, when the TCP throughput is relatively stable. The paper obtained bandwidth estimates differing by less than 10% when compared to running Iperf for 20 seconds, and savings in bandwidth estimation time of up to 94% and savings in network traffic of up to 92%.

Tags: TCP - IP

Jump to 1 [2] 3 4 5 6 7 8 9 10

Title	Date Added	Company
Group-Aware Stream Filtering for Bandwidth-Efficient Data Dissemination	2008-03-06	Dartmouth College
This paper is concerned with disseminating high-volume data streams to many simultaneous applications over a low-bandwidth wireless mesh network. For bandwidth efficiency, they propose a group-aware stream filtering approach, used in conjunction with multicasting, that exploits two overlooked, yet important, properties of these applications, many applications can tolerate some degree of "Slack" in their data quality requirements, and there may exist multiple subsets of the source data satisfying the quality needs of an application. This paper also analyzes the key factors that affect its performance, based on testing with heterogeneous filtering requirements. Tags: Data Quality
Humble ISD's Centralized, Clustered Network Infrastructure Is Improving the Quality of Education Through 20-Fold Faster Data Access, Greater Uptime, More Reliable Data Backup and a 40% Reduction in IT Administration Time	2008-03-01	Dell
The Humble Independent School District is named for a pioneering Texas settler, not for its approach to education. The fast-growing Humble Independent School District needed to improve network bandwidth and uptime, make data backup faster and more reliable, and reduce the time its IT staff spent on repairs and routine maintenance. Dell Services helped Humble ISD implement a highly robust, scalable server cluster, based on Dell PowerEdge servers. The cluster consolidates e-mail, office software and educational tools from 40 separate servers and offers dramatically faster access and simplified maintenance. The district turned its old data center into a mirrored disaster recovery site using Dell and EMC technology. Tags: Application Servers, Back-up
Microsoft Dynamics CRM 4.0: Performance and Scalability - Bandwidth Utilization Improvements	2008-03-01	Microsoft
Microsoft Dynamics CRM business software is designed to help enterprise organizations achieve a 360-degree view of their customers across marketing, sales, and service. Engineered to deliver performance that meets the needs of the largest global deployments, Microsoft Dynamics CRM has been tested for user scalability, data scalability, and network performance. This white paper focuses on network scalability improvements. Microsoft has made significant enhancements to the architecture of Microsoft Dynamics CRM 4.0 to improve network performance. Test results demonstrate network bandwidth utilization improvements of up to 94%. Improvements were evaluated in the number of round trips, bytes sent, and bytes received. Because the bandwidth utilization associated with a page can change after the first time the page is loaded, both initial (cold) and subsequent (warm) page loads were tested. Tags: Scalability, CRM Software, Network Design, High Performance Computing
A Novel Technique for SNMP Bandwidth Reduction: Simulation and Evaluation	2008-02-01	Taif University
One of the most popular and widely used Network Management Protocols (NMP) is SNMP. SNMP is extremely simple and versatile but it has scalability problems. A number of alternative NMPs such as RMI interfaces, CORBA and RMON have been attempted to overcome the problems with SNMP. However, these interfaces lack the standardization effort and simplicity of the usage that SNMP provides. This paper demonstrates a new technique to decrease a number of messages between manager and agent in the SNMP (i.e. traffic reduction). The base idea of the technique is to make a system manager keeps the MIB objects that are frequently required. They simulate the proposed technique using a network simulator called NS2. Tags: TCP - IP, Network Administration
Bandwidth Constrained IP Multicast Traffic Engineering Without MPLS Overlay	2008-01-01	University of Surrey
Existing multicast Traffic Engineering (TE) solutions tend to use explicit routing through MPLS tunnels. This paper shifts away from this overlay approach and address the bandwidth constrained IP multicast TE directly based on link state IGP routing protocols. The objective is that, through plain PIM-SM shortest path routing with optimized Multi-topology IS-IS (MISIS) link weights, the resulting multicast trees are geared towards minimal consumption of bandwidth resources. The paper applies Genetic Algorithms (GA) to the calculation of optimized M-ISIS link weights that specifically cater for engineered PIM-SM routing with bandwidth guarantees. Tags: Multicasting, MPLS
New Algorithm for Effective Utilization of Bandwidth for Sensitive Applications	2008-01-01	Indian Institute of Science
This paper proposes an algorithm, which allocates the bandwidth in an effective manner for sensitive applications. The main idea of this work is to allocate more amount of bandwidth for particular client, who has received more votes in the network. The proposed model reduces the wastage of bandwidth and increases the throughput. Tags: Software Engineering
ABwE: A Practical Approach to Available Bandwidth Estimation	2008-01-01	Stanford University
This paper reports on a new tool developed for monitoring available bandwidth in the range from several Mbps to 1000 Mbps. The tool is based on the packet pair dispersion technique. It has been in experimental use at SLAC for several months and it has been demonstrated at iGrid2002 and SC2002. It can be used for detecting bandwidth changes caused by routing or congestions. This paper has studied the influence of cross traffic and the behavior of routers on the delay of the probing packets. The paper presents part of these analyses together with results from monitoring of 12 paths in US, Japan and in Europe. Tags: TCP - IP
Achieving High Throughput in Low Multiplexed, High Bandwidth, High Delay Environments	2008-01-01	Columbia University
This paper explores combining ECN bits and loss feedback to achieve a high throughput connection in a high bandwidth, low multiplexed environment. In the scheme, a TCP source reacts differently on receiving an ECN than it does on inferring a packet loss in terms of congestion window update. Coupled with an AQM scheme such as RED, TCP sources achieve high utilization at a bottleneck router, but at the cost of fairness. The paper introduces a lightweight AQM mechanism that ensures fairness while still providing high utilization to the sources. The proposal is complementary to the HSTCP modification suggested by Floyd et al. Tags: TCP - IP
Trickle: A Userland Bandwidth Shaper for Unix-Like Systems	2008-01-01	monkey.org
As with any finite resource, it is often necessary to apply policies to the shared usage of network resources. Existing solutions typically implement this by employing traffic management in edge routers. However, users of smaller networks regularly find themselves in need of nothing more than ad-hoc rate limiting. Furthermore, the networks in question are typically unmanaged, with no network administrator(s) to manage complicated traffic management schemes. Trickle fills this gap by providing a simple and portable solution to rate limit the TCP connections of a given process or group of processes. Trickle works by taking advantage of the Unix dynamic loader's preloading functionality to interposition itself in front of the libc API interfaces that are used for sending and receiving data through BSD network sockets. Tags: TCP - IP, UNIX
Measuring End-to-End Bandwidth With Iperf Using Web100	2008-01-01	Stanford University
End-to-end bandwidth estimation tools like Iperf though fairly accurate are intrusive. This paper describes how with an instrumented TCP stack (Web100), one can estimate the end-to-end bandwidth accurately, while consuming significantly less network bandwidth and time. The paper modified Iperf to use Web100 to detect the end of slow-start and estimate the end-to-end bandwidth by measuring the amount of data sent for a short period (1 second) after the slow-start, when the TCP throughput is relatively stable. The paper obtained bandwidth estimates differing by less than 10% when compared to running Iperf for 20 seconds, and savings in bandwidth estimation time of up to 94% and savings in network traffic of up to 92%. Tags: TCP - IP