1. Characterize the customer's applications.
2. Characterize the network protocols.
3. Document the customer's current network.
4. Characterize Local and Non-Local Traffic.
5. Identify business constraints and inputs to your network design.
6. Characterize the existing network availability.
7. Characterize the network performance.
8. Characterize the existing network reliability.
9. Characterize the network utilizations.
10. Characterize the status of major routers.
11. Identify Potential Bottlenecks and System Vulnerability
12. Characterize the existing network management system and tools.
13. Summarize the health of the existing internetwork.

Now we will go into much more detail about each of these steps.

Step 1 - Customer Applications.  Use the table that follows the bulleted list to characterize the customer's applications, filling in the fields as indicated:

• In the Name of Application field enter the name of each application the customer runs over the network.
• In the Type of Application field enter information that will help you characterize the application such as database, multimedia, e-mail, scheduling, payroll, etceteras.
• In the Number of Users field enter the number of users who access each application.
• In the Number of Hosts field enter the number of hosts or servers that provide each application.
• In the Comments field add any relevant comments.  For example, you could write about scalability concerns, movements to migrate, special requirements on the network, etceteras.

The following table has been referred to above.

Step 2 - Network Protocols.  Use the table that follows the bulleted list to characterize the customer's network protocols, filling in the fields as indicated:

• In the Name of Protocol field enter the name of each protocol on the network.
• In the Type of Protocol field enter some additional information that will help you characterize the protocol such as client/server, peer-to-peer, connectionless, etceteras.
• In the Number of Users field enter the number of users who use each protocol.
• In the Number of Hosts field enter the number of hosts or servers that use each protocol.
• In the Comments field add any relevant comments.  For example, you could write about scalability concerns, movements to migrate, special requirements on the network, etceteras.

The following table has been referred to above.

Step 3 - Document Current Network.  You need to develop documentation on the customer's existing network.  Use their own documentation if reasonable.  While there are a lot of things that need to examined, the following list emphasizes some essentials:

• Network Topology -  draw a map or obtain a drawing from the customers of the network topology.  Include the type and speed of each major segment or link.  Also include names and addresses of major internetworking devices and servers.
• Addressing Schemes - document current addressing schemes.  Current addressing may well affect your ability to modify the network structure.  For example, a current IP subnet masking might limit your ability to add nodes in a LAN.
• Concerns About the Network - document any concerns you have about the existing network and any additional information that might not be apparent from the network topology map.  Characterize the overall network architecture to help you understand data flow patterns.

Step 4 - Characterize Local and Non-Local Traffic.  It is very important to identify potential bottlenecks associated with all sources.  It is usually best to use a protocol analyzer and determine how much traffic on each major network segment is local, not local and how much just passes through this segment.  When doing this it is important to identify source and destination addresses.  The following list describes some more about how you should use the table that follows after it.

• In the Network Segment Identification field enter an identifier for each segment.  This could be a logically related name or sequential numbering.
• In the Both Source and Destination Are Local field enter the percentage of traffic that satisfies this specified condition.
• In the Source Is Local, Destination Is Not Local field enter the percentage of traffic that satisfies this specified condition.
• In the Source Is Not Local, Destination Is Local field enter the percentage of traffic that satisfies this specified condition.
• In the Source Is Not Local, Destination Is Not Local field enter the percentage of traffic that satisfies this specified condition.

The following table is referred to above.

Step 5 - Identify Business Constraints and Inputs.  After you have interacted with your customer, input teams and future champions, check off as many of the following as you can.

	I understand the corporate structure.
	I have analyzed the information flow in the corporation.
	The customer has identified mission critical data and operations.
	The customer has explained any policies regarding approved vendors, protocols or platforms.
	The customer has explained any policies regarding open versus proprietary solutions.
	The customer has explained any policies regarding distributed authority for network design and implementation.  For example, departments that control their own internetworking purchases.
	I have a good understanding of the technical expertise of my clients.
	I have researched the customer's industry and competition.
	I am aware of any politics that might affect the network design proposal.
	I am aware of any financial constraints that may influence the network design.

 

You also want to make sure you document any concerns you have about the customer's business constraints.

Step 6 - Characterize Network Availability.  You need to make sure to gather statistics on network downtimes and MTBF - mean times between failures.  If some aspects or segments of the network are known to be fragile, gather statistics on those.  You also need to make sure to get the customer to express the cost of downtimes and their impacts on the overall business.

• What is the hourly cost to each department for a network outage?
• What is the hourly cost to each department for important particular outages on the network?
• What is the hourly cost to the overall organization for a network outage?
• What is the hourly cost to the overall organization for important particular outages on the network?

The following list describes some more about how you should use the table that follows after it.

• In the MTBF field enter the mean time between failures for each network segment or important component.
• In the Date of Last Downtime field enter the date on which the particular downtime was last experienced.
• In the Duration of Last Downtime field enter how long the last particular downtime lasted.
• In the Cause of Last Downtime field enter the cause, if known, of the last particular downtime.

The following table is referred to above.

Step 7 - Characterize Network Performance.  You need to make sure to gather statistics on network segments and devices to evaluate their performance.   This data should lead into later calculating utilizations.

The following list describes some more about how you should use the table that follows after it.

• In the Theoretical Capacity field enter the theoretical capacity of the segment or device.
• In the Typical Traffic Intensity field enter the traffic intensity that somehow represents typical activity on the segment or device.
• In the Peak Traffic Intensity field enter the traffic intensity for peak activity on the segment or device.
• In the Inward/Outward field enter the a ratio of inward to outward traffic for each component or device if reasonable.  This will often represent a ratio of downloads to uploads.

The following table is referred to above.

Step 8 - Characterize Network Reliability.  You need to make sure to gather statistics on network segments to evaluate their reliability.

The following list describes some more about how you should use the table that follows after it.

• In the Average Frame Size field enter the total number of MB transferred on the network or segment and divide it by the total number of frames.
• In the CRC Error Rate or Cyclic Redundancy Check Error Rate field enter the total number of CRC errors by the total MB transferred.
• In the MAC Layer Error Rate field enter the total number of MAC layer errors by the total number of frames.
• In the Broadcast/Multicast Rate field enter the total number of broadcasts/multicasts by the total number of frames.

The following table is referred to above.

. . .

Step 11 - Identify Major Bottlenecks and Vulnerabilities.  Now that you have all of the utilizations and performance measures for the routers, servers, and network segments/protocols you can look these over to determine where your network performance is most vulnerable and where your traffic flows relative to capacity are the above what you can tolerate in the system.

Step 12 - Characterize the Existing Network Management System and Tools.  Document the platforms and network management tools in use.  If available, gather examples of recent reports.

Step 13 - Summarize the Health of the Existing Network.  Check off any items on the following checklist that are true.  If the network is healthy you should be able to check everything.

Notice that the guidelines are quite vague and depend on many things like the type of traffic, applications, devices, topology, protocols, and criteria for acceptable performance.