Today's enterprise applications are demanding more and more network bandwidth. How do you ensure you have enough bandwidth and manage and it well? Find out how to handle bandwidth measurement and management with these resources.
Table of Contents
Bandwidth measurement introduction
Bandwidth is like the soul -- everyone thinks they know what you mean when you discuss it, but they are probably thinking about something entirely different. And that's because, like the soul, bandwidth does not transparently exist in any of the simple ways that we tend to believe.
Why begin this guide with bad metaphysics?
For a start, the word "bandwidth" is not even semantically sound -- it's a vague label improperly attached to a range of meanings (see
- bandwidth is a misnomer for more insight). Implicitly, it most typically refers to the maximum rate at which data can be transferred, either at a single interface or across an end-to-end path. But even this straightforward idea of "maximum data-transfer rate" can be endlessly qualified. Anything from packet size, to cross traffic, to corruptions can limit the effective transfer rate.
Somehow, though, bandwidth remains the central characterization for most networks. As if size matters. And it does.
For more information, read the "What is bandwidth?" definition from Whatis.com.
Bandwidth 101 will provide an introduction to bandwidth and reveal why people confuse bandwidth with "bit rate capacity." Learn how to measure bandwidth and why it is important to provide sufficient bandwidth for a signal.
Read bandwidth FAQs:
- Q&A: Which Layer 2 switching functions increase available bandwidth?
When switching at Layer 2, neither the router nor the switch are sent to the routing table. Rather, this refers to a cache, which stores the header information as well as the destination switch port. The device then begins switching at faster and faster speeds, making it appear as though there is more bandwidth at hand.
- Q&A: How can I determine the peak band width of my WAN connection?
There are free utilities that can check the actual peak bandwidth of a WAN connection in your enterprise network, including iNetCalc.
- Q&A: How do I allocate bandwidth dynamically in ad hoc network?
An ad hoc 802.11 network is by definition a peer-to-peer connection between two wireless stations. Those two stations will use all available bandwidth on the channel they have been configured to use (or have automatically chosen to use).
- Q&A: How do you interpret a bandwidth utilization graph?
It can be difficult to interpret whether bandwidth usage is "good, oversized or under-utilized," based solely on a bandwidth utilization graph. The best benchmark for measuring if a network link is over-utilized is to measure the impact of the resulting network bottleneck on the critical business applications and users using that link.
This tip on optimizing Internet bandwidth with link load balancing will reveal how link load balancing can provide an efficient, cost-effective and easy-to-use solution to maximize utilization of Internet access.
When calculating bandwidth, the behavior of user applications depends on the type, frequency of transaction, and other usage questions. Learn more about calculating bandwidth in this expert tip.
Looking to gather network traffic statistics and reports? Read about software that monitors a network-connected server and provides statistics about bandwidth use.
If the application is acting like the network is busy though you know it is empty, the network may be experiencing "phantom traffic" that s teals bandwidth.
TCP, or Transmission Control Protocol, often negatively impacts performance. There are a variety of relatively common situations that can lead to significant performance degradations due to TCP, or its interaction with other network behaviors:
- Wireless -- due to the inherently lossy nature of the medium, packet loss is typical and highly variable. TCP stacks such as Reno incorrectly infer that loss is congestion and slows down inappropriately. Bad cables, or other forms of malfunctioning media, can induce the same effect.
- Bandwidth/delay product -- very high latencies, for example network paths through satellites ( > 250ms), aggravate TCP's normal slow start mechanism, causing it to take very long times to reach full capacity. Once at full capacity, even a single lost packet can set back TCP's progress dramatically, particularly on high capacity links.
- Gigabit Ethernet -- poorly configured window sizes, or insufficient transmit or receive buffers, can impact the operation of TCP, restricting it unnecessarily or even causing it to fail. Tuning of the stack is possible although manual configuration is not a scalable solution.
- MTU black holes -- use of jumbo frames or badly implemented VPNs can cause MTU constrictions that fail to generate RFC1191-compliant error messages. Packets seem to disappear once the TCP window reaches a size greater than the hidden path MTU. Subsequently slow start and congestion avoidance interact cyclically with the black hole.
More on bandwidth management Learn questions you must ask your customers before deciding on bandwidth management services.
Find out how to educate customers about traffic management issues and offer bandwidth management solutions and service options.
Most (if not all) QoS mechanisms effectively limit bandwidth (or more properly, data transfer capacity). They do not ever increase it, certainly. They limit how much cross-traffic certain applications see -- at the expense of other applications. And those so-called premium applications will only ever see the network path as it might have appeared without QoS and no other traffic. This is a zero-sum game wherein anything gained for one application is lost elsewhere for others. Only when increasing bandwidth is not an option, technically or economically, should you turn to QoS for help. Even then, it won't be easy or trivially reliable, and it will inevitably cost you elsewhere.
NIC cards, or more specifically the drivers provided for them, won't necessarily perform at Layer 2 specification. In fact, the phrase "your mileage may vary" should be kept firmly in mind -- as well as the fact that there are many ways to improve your mileage and most of them are simple. Find out why NIC could be killing your network performance in this expert tip.
To improve your application performance, you need to identify the "narrowest" performance bottleneck that defines the end-to-end performance and remove it. And then remove the next one. And the one after that, until finally you achieve optimal network performance.
More on bandwidth management and measurement Find a list of free network monitoring tools that will help you keep your network in shape, without spending a dime.
Read an excerpt on bandwidth management from CCSP Self-Study: Cisco Secure Virtual Private Networks (CSVPN) to learn how you can configure the Concentrator to deliver the most effective bandwidth settings for your particular network.
Read an excerpt from Open Source Network Administration, including information on MRTG, which produces Web pages that display graphs of bandwidth use on network links on daily, weekly, monthly, and yearly scales.
Planning for bandwidth
In this tip on bandwidth management and bandwidth governance, learn why network managers have to do more than just manage bandwidth supply. They have to apply best governance practices to the consumption of bandwidth, so that utilization of network resources is closely aligned with business drivers. Only by exercising this kind of governance can IT use its infrastructure dollars in the most effective possible way.
For expert advice on bandwidth planning and new advancements in bandwidth technology, read the bandwidth challenge: How much is enough?
However, remember that more bandwidth does not always give you better performance.
Gigabit Ethernet (GigE) has rapidly gained prominence and acceptance as the next step in the evolution of corporate networks. The benefits of jumbo packets (>1500 bytes) on Gigabit Ethernets are significant. Today's network hosts are incapable of accessing the full Gigabit capacity with 1500 byte packets - jumbo packets can more than double accessible bandwidth on most machines. And yet there are some hidden perils. In the transition from 10 to 100 Mbps, issues with duplex mismatch and auto-negotiation became a serious headache on almost every network. In a similar way, MTU conflicts may hamper 100 Mbps to Gigabit transitions, and in some cases affecting performance more than duplex mismatches. But also like duplex mismatch, MTU conflicts are entirely avoidable with some careful planning.
For more information, read the definition of Gigabit Ethernetand find additional resources from WhatIs.com.
This was first published in February 2010