A Design Challenge – Part 1 with Virgilio SpazianiNetwork Design Methodology In Action: Challenge

 

On the Unleashing CCDE blog “Now What?”, a methodology was presented on how to create, select and justify network designs to meet elicited requirements. In this blog, I wanted to propose a challenge on how to apply that methodology as I suggest CCDE candidates and network designers in general who deal with complex, difficult-to-satisfy-all-together network requirements.

 

The goal of this challenge is not to test your IGP design skills, but rather to do the methodology walk-through to help you be in the network designer mindset and create network designs based on the requirements and within the constraints.

 

I describe here the steps to follow to try to solve this challenge, and in the next blog next week I will do the methodology walk-through to show you how to apply it in both real life situations and your approach to the CCDE Practical Exam.

 

Recommended Steps

0. Step “zero”:  know the technology

Of course the very first step – or better, the “zero” step – before starting to design is to know how the employed technologies work. Knowledge of IGP protocols and basic MPLS will be enough to understand and go deep in this challenge. The CCDE Written Exam’s Streamlined Preparation Resources spreadsheet (see link to the CCDE Written Exam Study Materials below) will help you to improve on your week points.

1. List all of the explicit and implicit business requirements in this challenge.

2. List all of the explicit and implicit technical requirements in this challenge.

3. List all of the explicit and implicit constraints in this challenge.

4. Compare and contrast side-by-side all options against the requirements and constraints

5. Justify the option chosen (why correct)

6. Justify the options not chosen (why incorrect)

 

Challenge

In this challenge you are the network designer for the fictitious Route First Service Provider in central Europe. You just came back from your holidays and started checking your emails when you found this email from the Director of Network Architecture:

 

To:network_designer@routefirst(dot)com.it

From: director@routefirst(dot)com.it

Subject: Migration to MPLS

Hi,

How are you? I hope you had a fantastic vacation, and now that you’ve come back from your holidays, it’s time to start designing the future of our company’s network.

As you know, our company is living in a period of deep transformation. Our success in Europe can make our future bright, but we’re facing some challenges in making our vision real. We really can be a leader in the advanced network solutions. We want a network that allows us to transport Internet data, video on demand, voice, and advanced services such as cloud computing and collaboration. We want to be a leader in the SP, advanced services space by proposing high quality network solutions to our customers at an aggressive price.

 

When we spoke about it before your vacations, we figured the best way to reach our goal is to build our network using technologies that will allow us in the future to better utilize and load share the network resources across different parts of our infrastructure. With such a network, we will reduce our need to increase circuit capacity, we will save money, and we will be very competitive from a cost prospective. We want to dedicate resources to gold customers and, if necessary, oversubscribe traffic from bronze customers.

 

Our company needs to migrate the current network to a MPLS Layer 3/Layer 2 VPN backbone. This network is currently based on multiple Layer 2 overlay VPN technologies. Because our field engineers are very experienced with OSPF, we want to continue to use OSPF on this expansion, unless we don’t have an alternative.

 

These are different zone types, part of the proposed MPLS backbone:

Core Zone: This is a 10-router network zone.

Aggregation Zone: This is a 10-router network zone.

Peripheral Zone: This is a 3-router network zone. The core routers will be part of the peripheral ring.

 

Based on those zones, we can build rings:

Aggregation rings: A pair of core routers and the associated aggregation zone routers will form an aggregation ring. At the moment, each ring is made by 12 routers: 10 aggregation and 2 core routers.

Peripheral rings: Each pair of aggregation ring routers (including both aggregation and core zone routers) and peripheral zone routers connecting to this couple will be in a 5-router peripheral ring (including the aggregation and core routers). The peripheral routers will not be replaced, and are low on memory and CPU (low-end routers).

 

Blog05-Figure1.png

We expect that in the future we will double the number of aggregation rings and aggregation routers per ring.

 

Options:

1. L2 IS-IS

Blog05-Option1.png

2. L1/L2 IS-IS

Blog05-Option2.png

3. Single area OSPF

Blog05-Option3.png

4. Multi area OSPF

Blog05-Option4.png

5. Multi area, multi process OSPF

Blog05-Option5.png

6. Multi area OSPF with Virtual Link

Blog05-Option6.png

7. EIGRP

 

Blog05-Option7.png

See the 7 design options for this challenge above. I need your expert recommendation on which option to use and why by next week.

 

Regards,

Mr. Route First Director of Network Architecture

 

avavirgilio2.png



Virgilio Spaziani is CCDE #20140003 and triple CCIE #35471 (R&S, SP, and Security). He’s a network designer and a Cisco official instructor based in Switzerland. He loves to solve complex network requirements using easy network designs, to teach complex technologies using easy examples.

 

 

 

Here are a few additional ways for us to engage and keep the conversation going: