External Merge Sort for Large Employee Records

A Java implementation of external merge sort for employee records that do not fit comfortably in memory. The project compares sorting criteria and writes reproducible CSV outputs for each result.

What it demonstrates

• External sorting over 16 input files containing 16,000 employee records.
• Sorting by employee ID, salary, department then salary, and first-name initial.
• Separation between domain models, file utilities, in-memory sorting, and external merge logic.
• A practical example of handling large datasets with bounded memory.

Technology

Java, CSV file processing, external merge sort, object-oriented design.

Run

Place the input CSV files in Data/, then compile and execute:

javac -d bin src/*.java
java -cp bin ExternalMergeSort

Sorted files are written back to Data/ using the criterion-specific filenames described in the source documentation.

Repository map

src/Employee.java defines the record model, EmployeeFileUtils.java handles file operations, InMemorySorter.java provides the baseline sorter, and ExternalMergeSort.java coordinates chunking and merging. See DATABASE.md and report.md for the design discussion.