Architecture Design Checklist

Describe Hardware and Software

A: If multiple hardware or software platforms are being considered for different sites, compare & contrast the features of each platform and explain why they will be deployed in different locations.

B: For each Stage of the project:

(1) List where and how many of each item (specific make and model, and the peripheral cards and cables and related equipment, software versions or support systems or scripts, and/or services) will be deployed. If the same hardware is used in locations of varying sizes, describe how the hardware will be sized at each location type. Describe how they will be connected to each other, and to the current underlying network (WAN links, LAN links, cables, interfaces, e.g. HSSI card to ATM card via 6 ft. RS-232 cable).

(2) Describe the routing topology (what's connected to what, and where). List the link and endpoint speeds and routing capabilities (locations that traffic can travel to from any point).

(3) Describe the underlying hardware and software environment that will support the added hardware and software (e.g. what location, space, power, temperature, and other environmental requirements exist; what OS, memory, application libraries, and other application requirements exist)?

(4) Describe the Network Naming conventions impacted by the new design: How should new equipment be named (by location, by Tier, by function)? How should new ports, routes, addresses, and other logical facilities be named?

Describe Required Vendor Service Changes

A: New services required, changes in service features or bandwidth. Determine from Cisco Contracts any implications of current vendor contracts on the changes under consideration.

Define SLA's and Architecture Implications

A: Capacity:

(1) What amount of traffic is anticipated from these new changes? (e.g. what size of user population is expected? How many transactions will be generated (per hour, day, or week)?) From where to where?

(2) What is the current capacity of the network at those locations, and what changes to the current architecture will need to be made in response to this new traffic level?

(3) What number of new hardware units, and what configuration, will need to be deployed to meet this capacity need?

(4) Lastly, what changes will need to be made to the architecture if traffic levels increase dramatically?

B: Availability:

(1) What availability levels are expected for this service? How does the architecture under consideration support those levels?

(2) Are there bottlenecks or single points of failure? Will backup routes, equipment, configuration backups, be needed to support those levels?

(3) In addition, what is the impact on current network availability by this change? What changes will be required in the current network architecture to ensure that current service availability will not be reduced?

(4) Lastly, if the availability of the new units in question is lower than anticipated, what architectural changes should be made to keep the overall network availability up? (e.g. diverse circuit routes, diverse circuit carriers, duplexed hardware options, hardware spare parts cached).

(5) Describe and the risk vs. cost tradeoffs for the recommended level of diversity. (e.g. "stocking an extra port board at every site will cost $165,000, and will result in a decrease in downtime from 15 hours per year to 5 hours per year at each location".)

C: Performance:

(1) What performance levels are expected for this service? (e.g. user response time, network response time, throughput, latency, user measured service quality. How does the architecture support overall network performance?

(2) Do the performance requirements vary with geography or type of office or location?

(2) Lastly, if the new equipment or software or management system performs at half the expected throughput or adds twice the expected latency, how would the architecture need to be changed to support the required network performance?

D: Service Quality:

(1) What other user-perceived quality metrics exist for this new service? (e.g. if voice service is involved, then P(dial tone), quality of voice tone, availability of PBX features;)

(2) If other metrics exist, how does the architecture support those quality levels?

8. Define Monitoring and Management Architecture:

A: How will these new network elements be monitored?

B: What parameters will be monitored, from where, and by who?

C: What parameters can be changed from a central remote management system, and when would this be used?

D: What remote management system will be used, and who will control it?

E: Are there databases (e.g. routing databases, call processing databases, configuration databases) that need to be linked to the new network elements for their proper operation? If so, what are they, and where will they be located, and how will they be connected to the new elements, and how will they be monitored and managed?

F: If there are multiple hardware or software platforms under consideration, how will they operate?

G: Provide a network diagram (e.g. a Visio diagram) showing the network locations and connections used for monitoring and management.

H: Are there expected changes to the way that current voice or data trouble reports are handled, or any changes needed to the current voice or data troubleshooting procedures? If so, what are these changes, and how should they be handled?

Security Architecture

A: Who will have superuser (read and write) access to all monitoring and management interfaces?

B: If users require special password access, where will the security database reside?

Configuration and Name Standards:

A: If IP addressable network elements are being considered for addition to the network, what is the plan for addressing these elements?

B: If routers or servers or other network equipment is being added, what is the plan for naming each of these devices?

Financial Analysis

A: Now that the architecture is well defined, revisit the earlier financial analysis done for the Project Plan. Compare the estimated benefits over time with the estimated costs (e.g. a table showing costs and benefits of current system, compared to the expected costs and benefits of the proposed new system). Determine the updated Return on Investment (ROI).