Close
Subscribe to rss feeds, which delivers links to the most popular articles, blogs via e-mail to your inbox every month

msn yahoo googletalk skype
  • News/Events

  • Statistics : Visits

    Online Visitor's Count: 1

    Last 24 Hours Count: 58

    Today's Count: 39

    Yesterday's Count: 71

    Last 30 Day's Count: 1684

  • Network Architecture and Design

    Problem Definition

    In today’s highly competitive and increasingly global marketplace, corporations are beginning to use their networks as a strategic advantage, one that empowers the business to gain an edge over their competitors. Networks are becoming a more visible, integral and essential part of a company’s business.

    IT organizations are being asked to evolve their existing network information systems to achieve lower operating costs, increased user access to critical business data, and greater flexibility in responding to an increasingly demanding business climate.  The end result is a steady stream of new technologies, constant pressure for new and improved network services, and increased requirements for higher bandwidths.

    Responding to these pressures in a piecemeal fashion, with multiple, compartmentalized purchases and frequent, uncoordinated changes in policy, will not keep the network ahead of the demand curve.  Neither will it keep the enterprise ahead of the competition.  Without a framework for guiding decisions on organization, policy, and new technology acquisition, building an efficient and effective communications capability is unlikely and the quality of the enterprise’s network will suffer.

    When a company relies on its enterprise-wide network as a critical business tool, it must carefully plan, design, implement, and manage the network, so that it supports the company’s business functions while servicing the individual needs of various organizations and end-users.

    The Networked Computing Practice, operating in Atlanta (USA), Kuala Lumpur (Malaysia), and Bangalore (India) was created to assist organizations in changing their enterprise’s approach to networking.

    The Networked Computing Practice consists of a virtual consulting team that leverages the skills and knowledge present in various businesses within IMS to provide three core network consulting services:

    • Network Strategy
    • Network Architecture
    • Network Design and Transition

     

    The Network Strategy and Architecture Service assists the client in defining a network strategy and a target network computing architecture.  The goal of network strategy is to make decisions on goals, priorities, and trade-offs for the network which reflect business priorities.  Choices are made about network technology flow from these decisions.  The network architecture is the target defined by the network strategy.  It provides a blueprint for network system purchasing and development that will minimize poorly targeted investments.  It allows an enterprise to consider the “big picture” and the total investment needed in the crucial area of information technology.  The transition strategy lays out the steps necessary to move the organization towards the network architecture.  These steps are in concert with the client’s business priorities.

    The Network Design and Transition Service develops a detailed network design based on the client’s future network computing architecture.  Design decisions and technology recommendations are aligned with the client’s business priorities, strategies and industry best practices.  The design approach standardizes on technologies which provide the highest cost benefits while positioning the network for integration of future information technologies.  The network design service provides the client with both a physical and logical design that considers the end-to-end communications components (session and transport layers) as well as the underlying networking infrastructure components (network protocols and physical topologies).

    Included with the design are the equipment specifications (functional requirements) needed to implement the new design.  Finally, the transition plan identifies the key implementation steps necessary for a client to move from their current network to the new network design.  These steps are aligned with the client’s network strategy and business priorities.

    Network Strategy

    The Network Strategy Service assists the client in defining a network strategy and a target networked computing architecture.  The goal of the network strategy is to make decisions on objectives, priorities, and trade-offs for the network which reflect business priorities.  Choices are made about network technology flow from these decisions.  The process followed in defining a network strategy is shown below:

    Figure 1 – Network Strategy Process

    In the first phase of developing the network strategy, IMS Consultants examine the client’s Business Strategy, IT Strategy, and the Existing Computing Architecture. The collected information is used to develop a baseline assessment of the network environment with regards to mission-critical applications, distribution of data, computing platforms, infrastructure, service levels, vendors, costs and demographics.

    The primary objective is to determine how the client interacts, or plans to interact, with its end-users, suppliers, clients, and/or partners using network and IT. The emphasis is on understanding the key business drivers with a focus on the enterprise’s mission-critical applications and the service level requirements.  Business needs to drive applications, which the network must support.

    The goal of the information-gathering phase is to obtain a clear understanding of the client’s priorities, strategic business initiatives being planned and future strategic directions where network and IT can contribute to the success of those efforts.  This phase involves a wide-ranging data collection effort which relies heavily on interviews but also draws from existing business strategy and technical documentation.  These findings will shape the Network Strategy materials.

    The Network Strategy Analysis uses a cross-organizational, interactive format to build consensus around the key drivers and criteria which will be used to shape the technology choices recommended by the network architecture.

    The value of the Analysis is that it develops a business-centered framework for assessing the technology solutions available to the enterprise over the next several years.  It provides a great deal of input in terms of what priorities, trade-offs, and principles are best for the enterprise.  Establishing these criteria is critical to linking the future network computing architecture to the business strategy.

    The final step in developing the overall Network Strategy is to document the results of the Network Strategy Analysis.  This is an analytical document which reports on decisions made about trade-offs and transitions, and provides a definition of the key guiding principles and evaluation criteria used to drive the network architecture process. The network strategy report includes documentation of the gaps between existing capabilities/resources and anticipated needs.  This analysis notifies the client of the most pressing gaps which may get in the way of achieving the network strategy.

     

    Network Architecture (High Level Design)

    The Networked Computing Architecture is the target defined by the Network Strategy.  It provides a long-term structure which identifies major system building blocks for the next generation internet. It specifies the interfaces, standards, and integration points at both the physical and logical level.  When correctly designed, the networked computing architecture brings unity, consistency, and simplicity to the design of a complex network.  It should also be cost effective, able to deliver the needed functionality, and open-ended enough to accommodate unforeseen requirements and new technologies.

    The architecture is considered a high-level design that provides the framework for the subsequent detailed design and implementation efforts.  This includes the end-to-end communications components (session and transport layers), and how the various computing platforms interface to the underlying network infrastructure.  It encompasses the connectivity, management, and interoperability needs at the local area network, building backbone, as well as the wide-area needs of site-to-site connectivity.  Inter-enterprise networking (Intranet) issues are also considered.  The transition plan defines a long-range target and includes the steps for implementing the full architecture, taking into consideration the client’s priorities.  The process followed in designing a networked computing architecture is shown in Figure 2.

     

    Figure 2 – Network Computing Architecture Process

    In the first phase of developing the Networked Computing Architecture, IMS Consultants examine the inputs from the client’s existing network strategy.  These inputs provide the definition of key guiding principles and evaluation criteria that will be used to drive the network architecture process.  This high-level analysis considers the existing business-critical applications and new/planned project initiatives.  The goal of this analysis is to establish an architectural approach that supports the client’s network strategy. The architecture must be able to deliver the connectivity, interoperability, manageability and performance required by the critical business application and new/planned project initiatives.  Because applications drive networking requirements, validation of the architecture approach is critical.

    The next step develops a set of Physical Network Alternatives for the networking infrastructure.  This includes the major system building blocks such as, sub-network technologies, network switching and interconnection devices, carrier services, and wiring systems. The most promising and applicable new network technologies are identified for the LAN, Campus, and Wide-Area Networks.  How each technology choice relates to the network strategy requirements for the physical infrastructure and applies to different sites and user types are also considered.

    The Logical Network Alternatives explores the technology choices associated with networking protocols and communication services in the application and end-user computing platforms.  The focus is on interoperability requirements, standard interfaces, legacy system support, terminal emulation and gate way strategies, the use of standards versus proprietary solutions, support for internet, extranet and intranet operability, etc.  The major technology choices, points of technical stability, and trade-offs are analyzed, including the associated strengths and weaknesses.  The mapping of each alternative versus the evaluation criteria and network principles are also described.

    During the Architecture Analysis, the physical and logical network alternatives, and the recommended direction is evaluated with the client.  The goal is to solicit client input on key architecture choices and to build consensus for the overall direction of the networked computing architecture.  The architecture direction is defined in the context of the key guiding principles and evaluation criteria identified in the network strategy.  The value of analysis is that it creates an analysis of the major options possible, and allows for informed trade-off discussions and decision making on the technology solutions available to the enterprise over the next several years.

    At this point, the major technology options, topologies, and standard interfaces have been selected and the focus of the next step is on documenting the details of the recommended network architecture.  The value of this document is that it defines the standards, structural guidelines and the recommended technology direction which the entire enterprise can gradually move towards.  By providing this framework, different groups and major business units can plan and implement changes to their local communications systems in a manner that will be consistent with the overall direction of the enterprise.

    The next step is to establish a framework for estimating the level of investment necessary to execute the architecture. The Cost Models estimate the anticipated capital investment needed to bring the enterprise into compliance with the architecture at a given point in time. It also estimates the on going operational and maintenance expenses for the enterprise network once the full architecture is in place.

    The final phase of the process is to develop and document a transition plan for realizing the network strategy and architecture over time.  This high-level plan outlines how the enterprise should implement the architecture to meet its network strategy objectives and realize the benefits of the proposed changes.  The value of this phase is that it presents an overall transition strategy and provides the client with a map of what it will take to make the transition to the recommended network computing architecture.  The Transition Strategy usually covers a period of three to five years, depending on the planning horizon of the client.


    Network Design (Low Level Design) and Transition (Migration)

    The Network Design and Transition Service develop a detailed network design based on the client’s target networked computing architecture.  Design decisions and implementation recommendations are aligned with the client’s business priorities, strategies and industry best practices.  The design process follows the standards, structural guidelines and recommended technologies identified by the Network Architecture.

    The network design service provides the client with both a physical and logical level design that considers the end-to-end communications components (session and transport layers) as well as the underlying networking infrastructure components (network protocols and physical topologies).

    Included with the design are the equipment specifications (functional requirements) needed to implement the new design.  Finally, the high-level implementation plan identifies the key steps necessary for a client to transition from their current network to the new network design.

    Figure 3 – Network Design Process

    The network design objectives are developed using the clients’ network strategy and future network computing architecture as input.  These objectives are defined in the context of the key guiding principles and evaluation criteria identified by the client. This establishes the measurement criteria and benchmarks for the final design to be provided by IMS.  (Output from the Network Strategy and Architecture Service can be used to provide these design objectives)

    The next step is to draft the Logical Level Design. This is accomplished by first building logical communication models of the existing or current computing environment.  The logical models document the session, transport and network layer interfaces used for end-to-end communications by each of the major applications. System profiles for both client and server platforms are then created to represent the end-to-end communications over the physical infrastructure. These logical models are built for each of the major applications supported by the network design.

    Once the communication models and platform profiles are built for the existing system, new models and profiles are developed to establish the communications architecture for new applications or future computing architecture.  The technology choices associated with networking protocols and communication services in the application and end-user computing platforms can be derived for the Network Architecture.  The Logical Design is one of the deliverables in the Network Design.  It specifies the overall approach, design criteria, design rationale, protocol recommendations and risks associated with a given design. It details the communication architecture by providing a complete “session-level through physical layer” catalogue of how each application communicates between the relevant client and server systems.

    After the logical level design is completed, the next step is to create the Physical Level Design.  First, we document the existing network topologies e.g. Building, Campus and Wide-Area Networks, the networking equipment/devices used in the various locations and the wide-area network, the networking equipment/devices used in the various locations and the wide-area network transmission facilities/services.  Next, we examine the type and size of the networked computing architecture.  We also investigate the in-country telecommunications environment for carrier service offerings at specified client locations and the estimated tariff costs.

    This input is then used by the designers to complete the draft physical level design for the networking infrastructure.  Detailed diagrams are created to show the overall structure and interconnection of the transport media with the network fabric.  This includes how major system building blocks such as, sub-network technologies, network switching and interconnection devices, carrier services, and wiring systems are applied to create the various Building, Campus and Wide-Area Networks for each of the different sites.  The approach to physical architecture and technology choices associated with the physical infrastructure can be derived from the Network Architecture.

    Certain clients may specify stringent requirements for network response times or data throughputs as part of the design criteria.  In these cases, the typical design service would be expanded to include a computer simulation of the draft design.  Traffic related information such as, messages/transaction types, source/destination, packet size, and arrival rates/times are entered into a network simulation tool used to model the design.  The modeling tool simulates the various transactions flows and traffic patterns across the network.  This provides the capability to more accurately estimate the utilization of the network facilities under normal operations and during error conditions such as, a link or backbone trunk failure.  This is especially useful in sizing complex wide-area networks.

    After the draft logical level design, the draft physical level design and the optional network simulation results have been reviewed with the client, the final designs are formalized into a design document.  Based on the final design, a set of Functional Requirements are established to specify the networking equipments and services needed to implement the new design.  This includes the equipment functionality, features, standards, interface types, port quantities, speeds, etc. that is required to support the design criteria.  This information can then be used by the client to establish their Tender or RFP specifications for the various network devices.

    The final step in the design process is to develop and document the Transition Plan. This is a high level plan that documents the recommended approach to implementation and outlines the necessary steps to remove any dependencies e.g. hardware, protocol and/or operating system dependencies.

    In addition to the Physical Design, Logical Design, and Transition Plan, the typical design document also includes a design support section(s) which describes our strategy for recovery from network faults, formulates the Network Test criteria and strategy for user acceptance, and describes the network management and operational controls that should be put in place to support the new network computing architecture.

     

    Summary

    Each of the Networked Computing Practice service offerings are delivered by a team of highly skilled and experienced consultants through proven processes and a consistent client engagement methodology.  During the service engagement, the client will be a partner and an active participant in the analysis, planning and design processes.  This ensures our on–going understanding of the client’s needs and provides one of the vehicles for transfer of knowledge.

    Each service offering is flexible in its design and can be easily customized to meet specific client needs and requirements.  Depending on the scope and deliverables outlined in the service offering, IMS Asia may draw upon additional resources such as, other IMS business units’ Consultants and Alliance Partners.