DBA Banking Dictionary: Essential Terms Every Database Administrator Should Know

Compact DBA Banking Dictionary: 100 Must-Know Terms for Banking SystemsManaging databases in banking requires both deep technical knowledge and a strong understanding of financial domain concepts. This compact dictionary collects 100 essential terms — technical, regulatory, and operational — that every DBA working with banking systems should know. Each entry includes a concise definition and, where helpful, a short note about why it matters to database administrators in banking.


1. ACID

Atomicity, Consistency, Isolation, Durability — properties guaranteeing reliable transaction processing. Critical for ensuring financial transactions are processed correctly and recoverably.

2. Active-Active

A high-availability configuration where multiple datacenters/processes actively serve traffic. Helps provide continuous availability and load distribution.

3. Active-Passive

One node serves traffic while another stands by for failover. Simpler than active-active but can have longer failover times.

4. Ad hoc reporting

On-the-fly queries and reports created by users. DBAs must control performance and security for such queries.

5. Aggregation

Combining data (e.g., sums, averages) for reporting or analytics. Important for OLAP workloads.

6. Audit trail

A log of actions and transactions for compliance and forensic purposes. DBAs must ensure immutability and retention.

7. Authentication

Verifying identity (users, services). Strong authentication reduces fraud and unauthorized access.

8. Authorization

Granting permissions to authenticated identities. Fine-grained authorization limits data exposure.

9. Availability zone

A physically separate location within a cloud region. Using multiple zones improves resilience.

10. Backup window

Scheduled time for backups. Must balance between minimal business disruption and data safety.

11. Batch processing

Processing large groups of transactions at scheduled times. Common for end-of-day banking jobs.

12. Big-endian / Little-endian

Byte order formats. Relevant when transferring binary data across systems.

13. Bistate data

Data representing two states (e.g., active/inactive). Simple, but must be handled carefully in workflows.

14. Blob (Binary Large Object)

Storage for large binary data (images, documents). Often used for scanned checks or statements.

15. Branch banking

Physical bank branches; systems must support branch-specific constraints and offline modes.

16. CAP theorem

Consistency, Availability, Partition tolerance — pick two in distributed systems. Guides DB design choices.

17. Cardinality

Number of distinct values in a column. Affects index choice and query plans.

18. Change data capture (CDC)

Tracking changes in data for replication or ETL. Useful for real-time analytics and auditing.

19. Checkpoint

A point where DBMS writes in-memory changes to durable storage. Important for recovery performance.

20. Clearinghouse

An entity that facilitates settlement between banks. Databases must integrate with clearing formats and schedules.

21. Clustering

Grouping servers/databases for availability or performance. Understand quorum and split-brain prevention.

22. Columnar storage

Data stored by column, optimized for analytics. Useful for reporting and fraud detection workloads.

23. Compliance

Adherence to laws/regulations (e.g., PCI DSS, GDPR). DBAs enforce policies on data handling and retention.

24. Consistency level

In distributed databases, degree to which reads reflect recent writes (e.g., strong, eventual). Choose based on transactional needs.

25. Contention

When multiple transactions compete for the same resources. Reducing contention improves throughput.

26. Contingency plan

Prepared response for catastrophic failures. Include recovery time objectives (RTO) and recovery point objectives (RPO).

27. Continuous integration / Continuous deployment (CI/CD)

Automated build and deploy pipelines. DB migrations must be integrated carefully.

28. Cross-site replication

Copying data across geographic sites. Important for disaster recovery and compliance.

29. Cryptographic hashing

One-way functions for data integrity and indexing (e.g., SHA-256). Used for checksums and fingerprinting.

30. Customer data file (CDF)

Central record of customer data. Strong protections and master-data management required.

31. Data anonymization

Removing personal identifiers for privacy-preserving analytics. Helps with regulatory compliance.

32. Data catalog

Metadata repository describing datasets. Helps analysts discover and trust data sources.

33. Data classification

Labeling data by sensitivity (e.g., public, confidential). Drives access controls and encryption policies.

34. Data governance

Processes and policies managing data quality, ownership, and lifecycle. DBAs implement technical controls.

35. Data lake

Storage for raw, large-scale data. Useful for ML and historical analysis in banking.

36. Data lineage

Traceability of where data came from and how it changed. Vital for audits and issue root-cause analysis.

37. Data masking

Replacing sensitive data with realistic but fake values for testing. Preserves privacy while enabling development.

38. Data model

Logical and physical schema design. Banking requires complex models for accounts, ledgers, and customers.

39. Data retention policy

Rules for how long data is kept. Must align with legal/regulatory requirements.

40. Data vault

A modeling technique for enterprise data warehouses emphasizing auditability and historical tracking.

41. Deadlock

Two+ transactions waiting indefinitely on each other. DBAs tune locking and isolation to prevent them.

42. Deduplication

Removing duplicate records or data blocks to save space. Useful for backups and storage efficiency.

43. Denormalization

Flattening data to improve read performance at cost of redundancy. Common in reporting systems.

44. Disaster recovery (DR)

Plans and systems to restore operations after major outages. DR drills are mandatory in banking.

45. Distributed ledger

A replicated, append-only ledger across multiple nodes (blockchain-like). Used in some modern payment systems.

46. Encryption at rest

Encrypting stored data. Required for protecting customer information.

47. Encryption in transit

Encrypting data while moving over networks (e.g., TLS). Prevents eavesdropping and tampering.

48. Event sourcing

Storing state changes as a sequence of events. Enables precise audit trails for transactions.

49. ETL (Extract, Transform, Load)

Pipelines that move and transform data into warehouses. Performance and correctness are critical.

50. Failover

Automatic switch to a standby system after failure. Test failovers regularly.

51. Fan-out

Sending a single event to multiple consumers. Useful in notifications and downstream processing.

52. Federation

Linking multiple databases under a unified access layer. Helps integrate legacy banking systems.

53. Flashback / Point-in-time recovery

Restoring database to a previous state. Useful for recovering from logical errors.

54. Foreign exchange settlement (FX settlement)

Processes for settling currency trades. Timing and reconciliation require precise database records.

55. Foreign key

Constraint enforcing relational integrity between tables. Ensures referential correctness.

56. GDPR (General Data Protection Regulation)

EU privacy law affecting data handling. DBAs must support subject access requests and deletion.

57. HA (High Availability)

Designs and practices to minimize downtime. Includes clustering, replication, and redundancy.

58. Hash partitioning

Distributing rows by hash of a key. Helps evenly spread load across shards.

59. Hot/warm/cold standby

Different recovery tiers for replicas (hot = ready-to-serve, cold = offline). Choose by RTO/RPO.

60. IAM (Identity and Access Management)

Systems controlling identities and permissions. Central to secure DB access.

61. Immutable ledger

An append-only log that resists modification. Useful for audit and compliance.

62. Index

A data structure to speed queries. Proper indexing dramatically affects performance.

63. Index fragmentation

When indexes become inefficient due to page splits and deletes. Periodic maintenance required.

64. In-memory database

Databases that keep working set in RAM for low-latency access. Useful for real-time fraud detection.

65. Input validation

Checking data correctness before accepting it. Prevents corruption and injection attacks.

66. Integration testing

Testing interactions between components. Necessary for migrations and upgrades.

67. Integrity constraints

Rules ensuring data validity (uniqueness, check constraints). Preserve correctness of financial records.

68. Isolation levels

Degree to which transactions are isolated (READ COMMITTED, SERIALIZABLE). Trade-offs between performance and consistency.

69. JSON/JSONB

Semi-structured data formats stored in DBs. Used for flexible payloads like customer preferences.

70. Key management

Handling of encryption keys. Secure storage and rotation are mandatory.

71. Key performance indicators (KPIs)

Metrics monitoring system health (latency, throughput). DBAs track KPIs to meet SLAs.

72. Ledger

Authoritative record of financial transactions. Must be tamper-evident and auditable.

73. Load balancing

Distributing work across servers. Reduce hotspots and improve performance.

74. Logging level

Granularity of logs (ERROR, INFO, DEBUG). Balance between observability and noise/storage.

75. Master data management (MDM)

Consistency of core business entities (customers, products). Prevents duplicate or divergent data.

76. Microsecond latency

Sub-millisecond response times relevant for high-frequency trading. Requires specialized infrastructure.

77. Middleware

Software connecting applications and databases. DBAs must understand how middleware affects transactions.

78. Mirroring

Maintaining real-time copies of data. Provides redundancy and improves read scalability.

79. Multi-tenancy

Single database serving multiple customers. Requires strict isolation and resource controls.

80. NAT (Network Address Translation)

Mapping private to public IPs. Relevant for network configuration of DB replicas.

81. Namespace

Logical grouping of database objects (schemas). Helps organize multi-application environments.

82. NoSQL

Non-relational databases optimized for flexibility or scale. Useful for certain banking workloads like session stores.

83. OLAP (Online Analytical Processing)

Systems optimized for complex queries and reporting. Separate from OLTP to avoid contention.

84. OLTP (Online Transaction Processing)

Systems optimized for transactional workloads (many small operations). Core for banking operations.

85. On-premises vs. cloud

Trade-offs between control and scalability. Many banks use hybrid approaches.

86. Operation window

Planned maintenance times. Communicate windows to stakeholders and minimize customer impact.

87. Orphaned records

Records that reference deleted parents. Detect and clean to maintain integrity.

88. Partitioning

Splitting tables/indexes to improve manageability and performance. Often by date for transaction tables.

89. PCI DSS (Payment Card Industry Data Security Standard)

Standards for protecting cardholder data. DBAs must ensure compliant storage and access controls.

90. Point-to-point encryption (P2PE)

Encrypting card data from entry to payment processor. Limits exposure inside systems.

91. PII (Personally Identifiable Information)

Data that can identify individuals. Requires strong protections and access controls.

92. Query planner / optimizer

Component that chooses execution plans. Statistics and indexes guide good plans.

93. Quorum

Minimum number of nodes required to make distributed decisions. Prevents split-brain and data divergence.

94. Rate limiting

Controlling request rates to protect systems. Helps prevent abuse and cascading failures.

95. Referential integrity

Ensuring relationships between tables remain consistent. Enforced with foreign keys and application logic.

96. Replication lag

Delay between primary and replica. Monitor lag for correctness of reads and failover safety.

97. Retry logic

Idempotent retries for transient failures. Design carefully to avoid duplicate financial effects.

98. Rolling upgrade

Upgrading nodes one at a time to avoid downtime. Useful for mission-critical systems.

99. Sharding

Horizontal partitioning across multiple servers. Needed for extreme scale of accounts or transactions.

100. Snapshot isolation

A concurrency control method using consistent snapshots for reads. Reduces read-write conflicts.


Security, compliance, and availability are the pillars of database work in banking. This compact list highlights the vocabulary DBAs need to design, operate, and secure systems that handle money, personal data, and regulatory obligations. Keep this dictionary handy and expand entries with platform-specific details (Oracle, PostgreSQL, SQL Server, Cassandra, etc.) as your environment requires.

Comments

Leave a Reply

Your email address will not be published. Required fields are marked *