PIG 文档搜索技能

# PIG Documentation Search Results: GROUP ALL Operator

## Overview
The `GROUP ALL` operator in Apache PIG aggregates all tuples in a relation into a single tuple. This is useful for operations like counting total records or calculating global statistics.

## Syntax
```
grouped_data = GROUP all_data ALL;
```

## Key Features
- **Single Output Tuple**: All input tuples are combined into one.
- **Use Cases**: 
  - Computing global aggregates (e.g., `SUM`, `COUNT`).
  - Filtering or transforming entire datasets uniformly.
- **Performance**: Efficient for small datasets but may strain memory for large ones.

## Example (from PIG 0.17 Docs)
```pig
-- Load data
raw_logs = LOAD 'web_logs' AS (user, url, timestamp);

-- Count total records
log_count = FOREACH (GROUP all_data ALL) GENERATE COUNT($1);
DUMP log_count;
-- Output: (15000)
```

## Best Practices
1. **Avoid on Large Data**: Use `GROUP` with `BY` for partitioned aggregation.
2. **Combine with `FOREACH`**: Apply transformations post-grouping.

*Source: Apache PIG 0.17 Documentation, [GROUP Operator](https://pig.apache.org/docs/r0.17.0/basic.html#GROUP)*