Table of Contents
Module 3 - Build a Prototype or Pilot for the Network Structure
Section 2 - Test the Prototype or Pilot
Section Objectives
Upon completion of this section, you will be able to:
-
List the Cisco IOS software commands you should use to determine if a network
structure meets the customer's performance and scalability goals.
-
Describe how to demonstrate the prototype or pilot to the customer so that
the customer understands that the proposed design meets requirements for
performance, security, capacity, and scalability, and that the costs and
risks are acceptable.
Time Required to Complete This Section
Approximately 1 hour
Completing This Section
Follow these steps to complete this section:
1. Study the reading assignment.
2. Click on any links that you see in the reading assignment and review
the information that appears.
3. Review any tables and job aids that appear in the reading assignment.
4. Review the case studies at the end of this section.
5. Complete the questions in each case study.
6. Review the answers provided by our internetworking experts.
Resources Required to Complete This Section
To complete this section, you will need:
-
Access to the World Wide Web and Cisco's Web site
-
A downloaded, printed copy of this section
-
Paper and pencil
Reading Assignment
Testing the Prototype or Pilot
Once you have installed and configured a pilot or a subset of the network
structure, you should test it using Cisco IOS software commands. If the
network is more sophisticated you may need to use a protocol analyzer.
For details on how to test with these tools and other simulation tools,
see the product documentation. Cisco IOS documentation is available at
the following link:
Documentation on the NETSYS network simulation tools is available at the
following links:
Using Protocol Analyzers
Configuring for Correct Media
To test with a protocol analyzer, make sure you have an analyzer that can
attach to the media you are testing, whether it is a WAN link, 10-Mbps
or 100-Mbps Ethernet, Token Ring, ATM, and so on. If necessary, configure
the analyzer for the media. For example, you might need to tell it whether
a Token Ring network is 4 Mbps or 16 Mbps. For WAN analyzers, you often
need to specify the data link layer, whether it is HDLC, Frame Relay, a
proprietary format, and so on.
Capturing Data
Assuming that the prototype network is installed and users are running
the appropriate applications, use the protocol analyzer to:
-
Capture data for at least a day.
-
Identify errors and irregularities.
-
Check the network utilization.
-
Determine what percentage of the traffic is broadcasts or multicasts.
To verify that there are no problems, use the Network Health Checklist
provided later in this section.
Testing the Prototype
You can also use a protocol analyzer to generate traffic when testing a
prototype. If it is impractical to purchase, install, and configure all
the devices required to do a true simulation, you can purchase a subset
of them and generate traffic to cause the load that would be present if
all the devices were installed. This simulation will give you an approximation
of the performance, though not an exact view of the actual performance
that you should expect.
Demonstrating Security
If the customer has security concerns and your prototype implements access
lists, use a protocol analyzer to demonstrate that traffic filtering is
working correctly. For example, if network A should not be able to reach
network B, run some network applications on hosts in network A while a
protocol analyzer is attached to network B. Prove to yourself and your
customer that no traffic from network A reaches network B.
Using Cisco IOS Software to Test a Prototype
In addition to using a protocol analyzer to test a prototype network, you
can use Cisco IOS software commands to make sure your network design will
work. The following commands are helpful when testing a prototype network
structure:
-
The show interface command identifies Layer 2 errors, router errors
such as dropped or ignored packets, and broadcast rates.
-
The show processes command identifies router CPU usage, including
CPU time used by processes.
-
The show buffers command checks buffer usage and buffer misses.
-
The ping and trace commands help troubleshoot connectivity
and performance problems.
-
The show [protocol] route command helps troubleshoot routing problems.
-
The show access-lists command displays the contents of access lists
to help troubleshoot security problems.
-
Various debug commands can be used for troubleshooting and verifying
packets sent and received.
Demonstrating Your Findings to the Customer
Once you have run some tests and proved to yourself that the network design
works, you will need to prove to the customer that the design works. Using
the results of your prototype tests, you should demonstrate that the design
meets the customer's requirements for performance, security, capacity,
and scalability, and that the design is within the customer's thresholds
for costs and risks. You can demonstrate your findings in many ways, including
the following:
-
Publish your findings in a concise but comprehensive report.
-
Add the results of your tests to the network design document or response
to an RFP that we discussed in Section 8 of Module
2, "Write a Design Document."
-
Make PowerPoint slides that graphically demonstrate the correlation between
your test results and the customer's requirements.
-
Meet with the customer and present your findings.
-
Meet with the customer and reproduce the test results as the customer watches.
Network Health Checklist
Use the following checklist when checking a prototype for proper network
functionality and scalability. This checklist provides guidelines, not
rules. As we mentioned in Module 1, the correct answer to questions regarding
thresholds for network health usually is "it depends." Thresholds depend
on topologies, router configurations, network applications, user requirements,
how measurement tools calculate thresholds, and many other factors.
Case Studies
In this section, you will build a prototype or recommend a pilot of the
network that you proposed in Module 1 and Module 2.
Read each case study and complete the questions that follow. Keep in
mind that there are potentially several correct answers to each question.
When you complete each question, you can refer to the solutions provided
by our internetworking experts. The case studies and solutions will help
prepare you for the Sylvan exam following the course.
In this section, you will review the following case studies:
1. CareTaker Publications, a publishing company
2. PH Network Services Corporation, a health care company
3. Pretty Paper Ltd., a European wall covering company
4. Jones, Jones, & Jones, an international law firm
Case Study: CareTaker Publications
To review the CareTaker Publications case study, click
here.
Recall the solution developed in the last section:
-
Because the network will be installed in a new building, we proposed that
the new communication lines be installed between the three locations in
parallel with the existing lines.
-
One of the new servers would be used as a test server. The server would
be placed on a 100-Mbps port of the switch.
-
A minimum of one PC would be attached to each LAN segment in the new building.
This attachment should be done in the computer room at the Catalyst 5000
switch to eliminate cable issues that are not the responsibility of the
internetwork design. A single PC at the warehouse would be connected to
the Cisco 2503 router. All circuits and paths would then be tested.
1. What criteria will you use to validate
that the pilot is a success?
2. Provide a list of the Cisco IOS commands you
will use to test your pilot network structure.
Now that you have completed the exercise, click here
to view the solutions provided by our internetworking design experts.
Case Study: PH
Network Services
To review the PH Network Services Corporation case study, click
here.
Recall the solution developed in the last section:
-
The critical path on the implementation of the PH Network system will be
obtaining circuits from the local provider.
-
The Frame Relay circuits should be ordered as soon as possible.
-
A structured plan should be established for doctor offices to place orders
for ISDN lines as a coordinated order for the entire network.
-
One hospital, one doctor office with an existing LAN, and one single-PC
doctor office should be selected as a Phase I Group. The communication
lines for these offices and the PH Network office should be installed first.
-
The Catalyst 1900 switch and Cisco 3620 router should be installed in the
PH Network office. Access to the Windows NT server by local PCs should
be verified.
-
The pilot network of one hospital, one LAN-based doctor office, and one
standalone-based office should then be connected and completely tested
by ping tests and user logins.
-
The new SQL Server could then be added and the connection tested for the
pilot network. Once these tests have proven the operational status of the
network, the other hospitals should be added, followed by doctor offices,
with an orderly, planned methodology.
1. What criteria will you use to validate
that the pilot is a success?
2. Provide a list of the Cisco IOS commands you
will use to test your prototype network structure.
Now that you have completed the exercise, click here
to view the solutions provided by our internetworking design experts.
Case Study: Pretty Paper Ltd.
To review the Pretty Paper case study, click
here.
Recall the solution developed in the last section:
-
The software vendor probably has been asked to provide this kind of proof
in the past. Ask if this vendor has a lab or test-bed facility that could
be used to simulate the environment if the routers and switch were provided.
The lab or test-bed facility would provide the most controllable environment.
-
If a test-bed is not available, all packet flow and performance requirements
can be evaluated by implementing a subset of the new network consisting
of a group of users in the main office and a remote manufacturing facility.
-
The pilot would consist of the installation of the Catalyst 5000 switch,
Cisco 2503 routers at headquarters and at a selected manufacturing facility,
the HP9000, two Frame Relay circuits, and the necessary subset of cabling.
-
Demonstrate that the solution enhances and does not degrade the flow of
packets from the warehouse to the main office. This demonstration would
be accomplished most definitively through the use of network monitoring
device for both the Frame Relay and the existing X.25 networks. At the
application level, this is not an apples-to-apples comparison. At the packet
transport level, it will be a comparison of the new network's ability to
deliver what it was asked to deliver.
1. What criteria will you use to validate
that the pilot is a success?
Now that you have completed the exercise, click here
to view the solutions provided by our internetworking design experts.
Case Study: Jones,
Jones, & Jones
To review the Jones, Jones, & Jones case study, click
here.
The firm decided to implement one U.S. office and the Europe office
as a prototype network. The main pilot office was located in the United
States. Both offices were fully cabled for the final design and the facilities
in both appropriately prepared.
1. What criteria will you use to validate
that the pilot is a success?
Now that you have completed the exercise, click here
to view an answer provided by our internetworking design experts.
CCDA Sample Sylvan Exam
We have provided a 32-question exam to simulate the Sylvan exam you
will be required to take to become CCDA certified. When you think you are
ready to try the sample CCDA exam, click here.
Copyright Cisco Systems, Inc. -- Version 2.0 7/98