System Integration Process, Lifecycle, and Multi-Application Fragmentation

Vocabulary flashcards covering core terms, lifecycle phases, deliverables, fragmentation challenges, and ETL pipeline concepts from the System Integration and Architecture (SIA301) module.

System Integration Lifecycle (SILC)

A structured six-phase framework followed by integrators to ensure enterprise system integrations are secure, sustainable, and reliable.

Requirements Gathering

phase in SILC focused on capturing technical, business, and operational goals, while defining databases, web services, manual file drops, latency, and traffic volume.

Analysis (SILC Phase)

phase in SILC where system analysts assess technical feasibility, translate business goals into concrete system needs, and identify potential issues such as incompatible databases or mismatched schemas.

Architecture Design

phase of the SILC where designers map the logical blueprint of an integration, detailing middleware platforms, security zones, data routing logic, and fallback mechanisms.

System Integration Design & Execution

phase of the SILC involving physical system integration, where engineers configure API gateways, map data schemas, establish message queues, and set up secure VPN tunnels.

Implementation & Verification

pahse of the SILC where the integrated system undergoes automated and manual testing, including regression, connection failure, schema validation, and end-to-end performance tests.

Maintenance (SILC Phase)

phase of the SILC involving ongoing post-deployment management, such as monitoring connection health, applying security patches, adapting to cloud API updates, and scaling resources.

Logical Architecture Diagram

A key deliverable produced during the Architecture Design phase that maps data pathways, security boundaries, and error-handling logic.

Application Fragmentation

An uncoordinated network of disparate enterprise software systems resulting from business units purchasing specialized software without a centralized IT strategy.

Data Drift

The occurrence of inconsistent data across systems when multiple applications store duplicate records of the same entity without a synchronization mechanism, causing records to diverge over time.

High human error rates

The average percentage range of human transcription errors in high-volume operations, which spans from $1\%$ to $5\%$.

Structural Staging Zones and ETL Data Pipelines

To resolve application fragmentation without risking direct database connections, developers use Extract, Transform, Load (ETL) pipelines and intermediate staging zones.

Extract (ETL Phase)

step in an ETL pipeline where raw data is pulled from a source system by querying database logs, subscribing to webhooks, or reading files.

Staging Zone

A secure, temporary storage layer in an integration pattern that holds extracted data before loading, preventing direct write operations, table locks, and data corruption on production databases.

Transform (ETL Phase)

The step in an ETL pipeline where raw data in the staging layer is cleansed, validated, converted, and reformatted to fit the target schema.

Load (ETL Phase)

step in an ETL pipeline where processed and formatted data is written into the destination database or delivered through a target API.

Single Source of Truth (SSOT)

An architectural governance policy where specific systems are given designated final authority over individual data fields across an enterprise network.