|  Infrastructure Patterns  |  Pattern Pictures  |  Impact Points  |  Architecture Components  |

Seven Core Infrastructure Patterns

Host (nondistributed). Internal employees performing scalable, batch, and highly deterministic OLTP from large sites and branches.

Two-Tier. Internal employees performing application functions that change rapidly (i.e., fast development cycles) from a local area network (not WAN or remote).   

Data Warehouse. Internal employees or partners performing broad-scope informational analysis (e.g., analytics, online database publishing) with local use of read-oriented data volumes from anywhere within the organization.

Enterprise Office. Internal employees using common applications with limited data sharing from every location within the organization.

Remote Access. Internal employees performing medium-volume transactional and local informational processing from remote locations.

Electronic Commerce. Employees, partners, customers, or external systems interacting with multiple system types (e.g., purchasing, inventory, etc. in both transactional and informational form) from anywhere in the world.

Impact Points

Availability: Defined as the impact of an applications outage on the business environment (suggested tiers: mission-critical, business-critical, important). This influences the network backup strategy, fault-tolerant needs, and network/systems management requirement (particularly event, fault, and health management).

Constituency requiring access: Defined as the individuals or resources requiring access to an application. This assists in “bounding” the application’s impact based on constituency economics, location, and activity level. It also communicates constituency-specific requirements (e.g., advanced authorization mechanism for remote uses) and illustrates constraints dictated by the scale of various constituency groups (e.g., “multiplier effect for small offices). Most important, by identifying the specific group of users, IT enables business people to decide the value of an infrastructure investment in business terms (i.e., whether a group of people is worth it). Organizations typically have to deal with six demographics: HQ/large sites, regional/branch offices, remote individuals, external individuals, external systems, and external content.

Data vulnerability: Defined as the relative level of information value and vulnerability. This assists in determining appropriate security requirements (e.g., internal firewall for network access control when systems are deemed mission-critical with sensitive data, strength of authentication, privacy requirements and encryption solutions.

Response time: Defined as the impact of latency (server or network) or bandwidth constraints on the applications (e.g., transaction times exceeding two or three seconds, batch jobs taking too long), traffic prioritization requirements, performance management tool needs, and the campus network (over)design must be examined.

Traffic patterns: Defined as the likelihood of significant changing traffic dynamics. For example, will the application be deployed to new sites, will large new sets of users be added, or will new functions be deployed soon? This helps determine contingency plans for upgrade scenarios and approximate implementation time frames.

Traffic volume (especially server, WAN): Defined as the relative (e.g., high, medium. Low) wide-area bandwidth requirement based on the constituency requiring access, application architecture, and traffic loads (e.g., query size, time of day, frequency). This assists in sizing WAN links and determining where workarounds are requi9red for applications that are not “network savvy” for non-local users. Also includes impact on device hardware/OS resources (e.g., server and desktop RAN and storage volumes.

Architecture Components

Application Technical:  Defined as non-hardware system components:  GUI, program-to-program communications, interoperability standards, transaction managers, messaging managers, Object Request Brokers, JAVA Beans.

Data Management & Movement:  Defined as method of data storage, retrieval, and management:  DBMS - Relational &, Object, ODBMS data storage technology, backup and recovery, data access technology (2-tier, and 3-tier), information access technology, query tools, DSS, OLAP, ROLAP, data warehousing, data marts, data mining, geographic information systems.  data migration, data replication, file transfer, bulk data movement, EDI, meta data.

Document Technology:  Defined as method for managing documents/content: Document format (e.g. HTML, XML), creation tools, document repositories, collaboration, document management systems, content management.

Network Services:  Defined as services integral to the network but not network hardware: File services, print services, electronic mail, discussion databases, workflow/groupware, network addressing, address assignment, domain name services, directory services, time synchronization, desktop conferencing, web servers, wireless application protocols, streaming media, video conferencing, (DHCP, WAP, HTTP, IIOP, LDAP, X.509, SSL, MAPI, IMAP, SMTP, MIMI, SMILE, POP)

Networks:  Defined as physical and fundamental software components upon which Network Services reside: Protocols, topology, WAN, MAN, LAN architectures, backbones technology, speeds, virtual private networks, switching layers,  routers/switches, routing protocols, 

Network. & Systems Management:  Defined as the methods and tools that manage the operation and performance of all computing and network components.  Network configuration and management, platform configuration and management, trouble/incident management.

Platforms:  Defined as physical computing platforms:  Mainframes, mid-range, and small computers, network servers, workstations, terminal servers, hand-held devices, warehouse hardware, printers, scanners.

Security:  Defined as the methods and technology for managing security:  Physical security, directory for uses profiles, identification, authentication, authorization, remote access methods, biometric identification, digital certificates, virus detection, firewalls, DMZ's, intrusion detection, vulnerability assessment.

Development:  Defined as the methods and technology for managing software:  Development platforms, test platforms, operational platforms, development methodology, project management, requirements management, function point management, data modeling, reverse engineering, object oriented modeling, configuration management, source code management, GUI standards, programming languages, naming conventions, integrated development environments, software quality assurance, component and system testing, regression testing.

Telephony (CTI):  Defined as the methods and technology for integrating telephone technology:  Integrated Voice Response, automated dialing, predictive dialing, automated fax, screen pops, call routing, call management, call monitoring, reporting, automatic number identification.