HintsToday

💡 What is Teradata?

Teradata is a massively parallel processing (MPP), enterprise-grade relational database management system (RDBMS) designed specifically for data warehousing and large-scale analytics.

🏗️ Core Purpose

Teradata is built to store and analyze massive volumes of structured data efficiently.

It helps large organizations—especially banks, telecoms, and retailers—run:

• Complex SQL analytics
• Fast reporting and BI
• Data mining, predictive modeling
• Regulatory and compliance reporting

🔧 Key Features of Teradata

🔐 Typical Banking Use Cases

🛠️ Tools & Technologies with Teradata

Languages

• SQL (ANSI standard)
• BTEQ (Basic Teradata Query) – scripting tool for batch execution
• Teradata SQL Assistant – GUI SQL interface
• FastLoad, MultiLoad, TPT – bulk data load utilities

Tools

📦 Teradata Architecture (Simplified)

 +----------------+       +------------------+
 | Client (SQL)   | <-->  |  Parsing Engine  |  <-- manages SQL, session
 +----------------+       +------------------+
                                 |
                         +---------------+
                         | BYNET Network |
                         +---------------+
                                 |
              +-------------------+-------------------+
              |                   |                   |
        +----------+        +----------+        +----------+
        | AMP Node |        | AMP Node |  ...   | AMP Node |
        +----------+        +----------+        +----------+
        | Storage  |        | Storage  |        | Storage  |

• AMP = Access Module Processor → Each handles a portion of data and processes independently.
• BYNET = Teradata’s communication layer between nodes

🏁 Summary

Absolutely! Let’s dive into a step-by-step walkthrough of how data flows in a banking ecosystem — covering OLTP databases, Data Lakes, Data Warehouses, and analytics use cases.

🏦 1. Where Daily Banking Transactions Are Stored

✅ Source Systems: OLTP Databases

• Systems like: Core Banking, Loan Management, ATM Systems, Credit Card Systems, Mobile App Backend, CRM, etc.
• Databases Used: Oracle, IBM Db2, PostgreSQL, SQL Server, MySQL
• Characteristics:
  • High speed, low latency
  • Row-based storage
  • Handles millions of inserts/updates per day
  • Highly normalized schema
  • Not suitable for heavy analytical queries

📝 Examples of Data Stored Here:

📤 2. ETL or ELT to Central Storage

At end of day (EOD) or in real-time (for some events), data is:

• Extracted from OLTP systems
• Transformed: cleaned, enriched, joined
• Loaded into:
  • Data Lake (for raw + semi-structured/unstructured)
  • or Data Warehouse (for structured, analytical-ready)

🗂️ 3. Data Lake in Banking: Use & Architecture

✅ When Data Lake is Used:

• For storing raw, unstructured, semi-structured, and structured data
• To keep history of all ingested data
• For data science, ML, sandbox experiments
• For regulatory traceability (raw audit logs)

🔧 Storage Technology:

• On-Prem: Hadoop (HDFS)
• Cloud: Azure Data Lake, Amazon S3, Google Cloud Storage

📁 Typical Zones in a Data Lake:

📝 Data Formats:

• CSV, JSON, Avro, Parquet
• Images, PDFs (e.g., cheque scans), logs

📦 Examples:

• Log files from ATMs
• Mobile app clickstream events
• OCR from cheque images
• Web scraping for credit intelligence

🏛️ 4. Data Warehouse in Banking

✅ When Data Warehouse is Used:

• For business reporting, dashboards, regulatory reporting
• For fast SQL queries across large volumes of data
• When data is cleaned, modeled, and schema is defined
• Used by Finance, Risk, Compliance, BI Teams

🔧 DW Technologies:

• On-Prem: Teradata, Netezza, Oracle Exadata
• Cloud: Snowflake, Redshift, Azure Synapse, BigQuery

🧱 Typical Design:

• Star / Snowflake Schema
• Facts + Dimensions (e.g., transactions fact, account dimension)

🔁 Data Flow:

Daily Transactions → ETL → Data Warehouse (facts/dims) → Power BI / Tableau dashboards

📝 Example Queries:

• “Show top 100 customers by loan amount”
• “Find accounts inactive for 180 days”
• “Credit card spend trend by customer segment”

🔄 5. Data Flow Overview in a Bank

+-------------+        +-----------+        +-----------+        +------------+
| OLTP System | -----> | ETL Tools | -----> | Data Lake | -----> | Data Warehouse |
+-------------+        +-----------+        +-----------+        +------------+
   (Oracle,          (Informatica, Talend)   (Azure ADLS, S3)     (Snowflake, Teradata)
    SQL Server)

       ↓                                       ↑            ↑
 Real-Time Kafka → Stream Processing (Spark, Flink) →      |
       ↓                                                    |
      ML Models, AI Apps, Risk Engine                     BI Tools

🧠 Common Tools in Each Layer

📌 When to Use What?

✅ Summary: Roles of Databases, Lakes, and Warehouses in Banking

Pages: 1 2 3 4