XML Big Data Analytics
Scenario: Tổng hợp doanh số từ 50TB hóa đơn XML trong 10 năm
Data Size
50 TB
Cluster Nodes
1,000
Total Blocks
~400K
HDFS Ingestion
Raw XML Data (Sample)
<invoice> <id>INV_2014_001</id> <date>2014-01-15</date> <items> <item> <product_id>SP001</product_id> <quantity>10</quantity> </item> <item> <product_id>SP002</product_id> <quantity>5</quantity> </item> </items> </invoice> <invoice> <id>INV_2015_099</id> <date>2015-06-20</date> <items> <item> <product_id>SP001</product_id> <quantity>20</quantity> </item> <item> <product_id>SP003</product_id> <quantity>8</quantity> </item> </items> </invoice> <invoice> <id>INV_2023_555</id> <date>2023-12-01</date> <items> <item> <product_id>SP002</product_id> <quantity>15</quantity> </item> <item> <product_id>SP001</product_id> <quantity>5</quantity> </item> </items> </invoice>
Lưu vào Data Lake (Phân tán)
HDFS Distribution (Simulated 1000 Nodes)
Waiting for data...
Cluster Storage Empty
Processing
Spark/MapReduce Job Execution
Chạy Job Tổng Hợp
Giai đoạn MAP (Parse XML)
1000 Mappers chạy song song
Waiting for job submission...
SHUFFLE & REDUCE (Tổng hợp)
Aggregation by Product_ID
Waiting for mappers...
Kết quả phân tích (Final Output)