{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "benchmarks": {
    "overview": {
      "testScenario": "Export 15 million rows × 9 columns to CSV format",
      "dataVolume": "135 million cells",
      "metric": "Cells per second",
      "hardwareSpecs": "Standard enterprise server configuration",
      "datePerformed": "2024"
    },
    "results": [
      {
        "tool": "FastBCP",
        "vendor": "ARPE",
        "category": "Specialized Database Export Tool",
        "performance": {
          "cellsPerSecond": 24000000,
          "formattedSpeed": "24M cells/s",
          "rank": 1
        },
        "speedMultiplier": {
          "vsNativeTools": "8.6x faster",
          "vsSsis": "13.3x faster",
          "vsInformatica": "20x faster",
          "vsSapBods": "26.7x faster",
          "vsTalend": "32x faster"
        },
        "estimatedTimeFor100GB": "~8 minutes",
        "highlighted": true
      },
      {
        "tool": "Native Database Tools",
        "examples": ["BCP (SQL Server)", "Sqlldr (Oracle)", "COPY (PostgreSQL)"],
        "category": "Database Native Utilities",
        "performance": {
          "cellsPerSecond": 2800000,
          "formattedSpeed": "2.8M cells/s",
          "rank": 2
        },
        "estimatedTimeFor100GB": "~71 minutes",
        "notes": "Optimized for single-threaded operations"
      },
      {
        "tool": "SSIS",
        "vendor": "Microsoft",
        "category": "Enterprise ETL Platform",
        "performance": {
          "cellsPerSecond": 1800000,
          "formattedSpeed": "1.8M cells/s",
          "rank": 3
        },
        "estimatedTimeFor100GB": "~111 minutes"
      },
      {
        "tool": "Informatica PowerCenter",
        "vendor": "Informatica",
        "category": "Enterprise ETL Platform",
        "performance": {
          "cellsPerSecond": 1200000,
          "formattedSpeed": "1.2M cells/s",
          "rank": 4
        },
        "estimatedTimeFor100GB": "~167 minutes"
      },
      {
        "tool": "SAP BODS",
        "vendor": "SAP",
        "category": "Enterprise ETL Platform",
        "performance": {
          "cellsPerSecond": 900000,
          "formattedSpeed": "900K cells/s",
          "rank": 5
        },
        "estimatedTimeFor100GB": "~222 minutes"
      },
      {
        "tool": "Talend",
        "vendor": "Talend",
        "category": "Open Source ETL Platform",
        "performance": {
          "cellsPerSecond": 750000,
          "formattedSpeed": "750K cells/s",
          "rank": 6
        },
        "estimatedTimeFor100GB": "~267 minutes"
      }
    ],
    "performanceSummary": {
      "fastBCPAdvantage": {
        "averageSpeedup": "16x faster than competing ETL tools",
        "maxSpeedup": "32x faster than slowest competitor",
        "vsNativeTools": "8.6x faster than database native tools"
      },
      "timeComparison100GB": {
        "fastBCP": "~8 minutes",
        "competitorAverage": "~168 minutes",
        "timeSaved": "~160 minutes (95% faster)"
      }
    },
    "realWorldExamples": [
      {
        "scenario": "Oracle to Azure Data Lake (Parquet)",
        "dataSize": "500GB table",
        "fastBCPTime": "~40 minutes",
        "traditionalToolTime": "~10.6 hours",
        "timeSaved": "~10 hours",
        "savingsPercent": 94
      },
      {
        "scenario": "SQL Server to AWS S3 (CSV)",
        "dataSize": "50 million rows, 25 columns",
        "fastBCPTime": "~6 minutes",
        "traditionalToolTime": "~90 minutes",
        "timeSaved": "~84 minutes",
        "savingsPercent": 93
      },
      {
        "scenario": "PostgreSQL to GCS (JSON)",
        "dataSize": "10 million rows, 50 columns",
        "fastBCPTime": "~4 minutes",
        "traditionalToolTime": "~65 minutes",
        "timeSaved": "~61 minutes",
        "savingsPercent": 94
      }
    ],
    "performanceFactors": {
      "parallelization": {
        "description": "Multi-threaded data extraction and processing",
        "impact": "Primary performance driver - scales with CPU cores"
      },
      "streaming": {
        "description": "Memory-efficient streaming architecture",
        "impact": "Eliminates memory bottlenecks, enables large dataset processing"
      },
      "nativeDrivers": {
        "description": "Optimized database drivers embedded in executable",
        "impact": "Reduced overhead compared to ODBC/JDBC layers"
      },
      "directCloudUpload": {
        "description": "Direct streaming to cloud storage",
        "impact": "Eliminates local disk I/O bottleneck"
      },
      "optimizedCodec": {
        "description": "Efficient encoding for Parquet and compression",
        "impact": "Faster file writing with smaller output size"
      }
    }
  },
  "metadata": {
    "lastUpdated": "2026-02-19",
    "purpose": "Performance benchmarks for AI agents",
    "schemaVersion": "1.0.0",
    "disclaimer": "Benchmark results may vary based on hardware, network, database configuration, and data characteristics. These results represent typical performance under standard conditions."
  }
}