Product Support
Product Activation FAQs
Downloads
Maintenance Programs
White Papers
WildPackets Forums
Technical Compendium
Additional Resources
Tip of the Month
Wireless Tips
Plugin Tips
Product Versions
Networking Books
Networking Glossary
Networking Links
Tech Support Requests
Home > Support > Additional Resources > Tip of the Month

Tip of the Month

July 2004

The Network is slow or the server?

The first assumption that we make is that we're dealing with a protocol in which a command is sent and a response is received. It's possible to apply this technique to protocols that don't respond to the command (e.g. broadcast-only protocols), but command/response protocols are where users typically feel network slowdown the most. Measuring latency--the amount of time between the transmission of the command and the receipt of the response--is the first step in troubleshooting network slowdown. If latency is high, the user will perceive slow network performance.

Latency in a command/response protocol consists of three components: outbound latency, processing delay, and inbound latency.

Outbound latency is the elapsed time from the transmission of the command (by the client) to its reception (by the server). Inbound latency is the elapsed time from the transmission of the response (by the server) to its reception (by the client). Processing delay is the time between the reception of the command (by the server) and the transmission of the response (by the server).

The question "Is the network slow or is the server slow?" can be answered by comparing the inbound and outbound latency to the processing delay. If inbound or outbound latency are excessively long, then the network is slow. If processing delay is excessively long, then the server is slow. These three components can be measured by capturing a command/ response pair and measuring the delta time. If EtherPeek is located close to the client, the elapsed time between the command/response pair represents the total round-trip latency, including all three components. If EtherPeek is located close to the server, the elapsed time between the command/response pair represents only processing delay. Total latency minus processing delay equals network latency (inbound + outbound).

In reality, evaluating latency is only the first step in answering the "slow network" question. Many factors other than high latency can cause the user to perceive that the network is slow. Evaluation of latency should be the first step, however. A network with low latency might still feel slow, but a network with high latency probably won't feel fast. If your evaluation of latency reveals that the processing delay is the cause of the users' complaints, then upgrading the server, not the network, might be the answer.

Download a demo of OmniPeek
Custom Integration & Engineering
WildPackets understands that one size does not fit all. Moreover, we all face new challenges every day. WildPackets Custom Engineering performs software development and systems integration, complementing WildPackets products and enhancing the capabilities of Network Operations Centers.
Learn more...

Network Analysis & Consulting
WildPackets offers a full spectrum of professional services, available remote and on-site. Our network engineers provide expertise for your network troubleshooting, capacity planning, or baseline performance analysis needs.
Learn more...

Tip of the Month
Pump Up The Volume
If you are using WildPackets® OmniEngine or WildPackets® Omnipliance, chances are you are interested in collecting, storing, and analyzing very large volumes of packet data. The key word here is VOLUME! On highly utilized gigabit or 10 gigabit links, hundreds of thousands, or even millions, of packets can be collected in just a few seconds. Processing that many packets takes a lot of horsepower, and OmniEngine is strong enough to handle the flow in most cases. However, every computer system has its limits, so you need to know how to maximize the packet volume that OmniEngine can accommodate. Here are some best practices to consider when setting up your capture options.