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.
Leave a Reply