IMPORTANT: No additional bug fixes or documentation updates will be released for this version. For the latest information, see the current release documentation.
Elastic Docs ›Elasticsearch Guide [7.11] ›Roll up or transform your data ›Transforming data
« Tutorial: Transforming the eCommerce sample data Painless examples for transforms »

Transform examplesedit

These examples demonstrate how to use transforms to derive useful insights from your data. All the examples use one of the Kibana sample datasets. For a more detailed, step-by-step example, see Tutorial: Transforming the eCommerce sample data.

Finding your best customersedit

This example uses the eCommerce orders sample data set to find the customers who spent the most in a hypothetical webshop. Let’s transform the data such that the destination index contains the number of orders, the total price of the orders, the amount of unique products and the average price per order, and the total amount of ordered products for each customer.

POST _transform/_preview
{
  "source": {
    "index": "kibana_sample_data_ecommerce"
  },
  "dest" : { 
    "index" : "sample_ecommerce_orders_by_customer"
  },
  "pivot": {
    "group_by": { 
      "user": { "terms": { "field": "user" }},
      "customer_id": { "terms": { "field": "customer_id" }}
    },
    "aggregations": {
      "order_count": { "value_count": { "field": "order_id" }},
      "total_order_amt": { "sum": { "field": "taxful_total_price" }},
      "avg_amt_per_order": { "avg": { "field": "taxful_total_price" }},
      "avg_unique_products_per_order": { "avg": { "field": "total_unique_products" }},
      "total_unique_products": { "cardinality": { "field": "products.product_id" }}
    }
  }
}

The destination index for the transform. It is ignored by _preview.

Two group_by fields is selected. This means the transform contains a unique row per user and customer_id combination. Within this data set, both these fields are unique. By including both in the transform, it gives more context to the final results.

In the example above, condensed JSON formatting is used for easier readability of the pivot object.

The preview transforms API enables you to see the layout of the transform in advance, populated with some sample values. For example:

{
  "preview" : [
    {
      "total_order_amt" : 3946.9765625,
      "order_count" : 59.0,
      "total_unique_products" : 116.0,
      "avg_unique_products_per_order" : 2.0,
      "customer_id" : "10",
      "user" : "recip",
      "avg_amt_per_order" : 66.89790783898304
    },
    ...
    ]
  }

This transform makes it easier to answer questions such as:

  • Which customers spend the most?
  • Which customers spend the most per order?
  • Which customers order most often?
  • Which customers ordered the least number of different products?

It’s possible to answer these questions using aggregations alone, however transforms allow us to persist this data as a customer centric index. This enables us to analyze data at scale and gives more flexibility to explore and navigate data from a customer centric perspective. In some cases, it can even make creating visualizations much simpler.

Finding air carriers with the most delaysedit

This example uses the Flights sample data set to find out which air carrier had the most delays. First, filter the source data such that it excludes all the cancelled flights by using a query filter. Then transform the data to contain the distinct number of flights, the sum of delayed minutes, and the sum of the flight minutes by air carrier. Finally, use a bucket_script to determine what percentage of the flight time was actually delay.

POST _transform/_preview
{
  "source": {
    "index": "kibana_sample_data_flights",
    "query": { 
      "bool": {
        "filter": [
          { "term":  { "Cancelled": false } }
        ]
      }
    }
  },
  "dest" : { 
    "index" : "sample_flight_delays_by_carrier"
  },
  "pivot": {
    "group_by": { 
      "carrier": { "terms": { "field": "Carrier" }}
    },
    "aggregations": {
      "flights_count": { "value_count": { "field": "FlightNum" }},
      "delay_mins_total": { "sum": { "field": "FlightDelayMin" }},
      "flight_mins_total": { "sum": { "field": "FlightTimeMin" }},
      "delay_time_percentage": { 
        "bucket_script": {
          "buckets_path": {
            "delay_time": "delay_mins_total.value",
            "flight_time": "flight_mins_total.value"
          },
          "script": "(params.delay_time / params.flight_time) * 100"
        }
      }
    }
  }
}

Filter the source data to select only flights that are not cancelled.

The destination index for the transform. It is ignored by _preview.

The data is grouped by the Carrier field which contains the airline name.

This bucket_script performs calculations on the results that are returned by the aggregation. In this particular example, it calculates what percentage of travel time was taken up by delays.

The preview shows you that the new index would contain data like this for each carrier:

{
  "preview" : [
    {
      "carrier" : "ES-Air",
      "flights_count" : 2802.0,
      "flight_mins_total" : 1436927.5130677223,
      "delay_time_percentage" : 9.335543983955839,
      "delay_mins_total" : 134145.0
    },
    ...
  ]
}

This transform makes it easier to answer questions such as:

  • Which air carrier has the most delays as a percentage of flight time?

This data is fictional and does not reflect actual delays or flight stats for any of the featured destination or origin airports.

Finding suspicious client IPsedit

This example uses the web log sample data set to identify suspicious client IPs. It transform the data such that the new index contains the sum of bytes and the number of distinct URLs, agents, incoming requests by location, and geographic destinations for each client IP. It also uses filter aggregations to count the specific types of HTTP responses that each client IP receives. Ultimately, the example below transforms web log data into an entity centric index where the entity is clientip.

PUT _transform/suspicious_client_ips
{
  "source": {
    "index": "kibana_sample_data_logs"
  },
  "dest" : { 
    "index" : "sample_weblogs_by_clientip"
  },
  "sync" : { 
    "time": {
      "field": "timestamp",
      "delay": "60s"
    }
  },
  "pivot": {
    "group_by": {  
      "clientip": { "terms": { "field": "clientip" } }
      },
    "aggregations": {
      "url_dc": { "cardinality": { "field": "url.keyword" }},
      "bytes_sum": { "sum": { "field": "bytes" }},
      "geo.src_dc": { "cardinality": { "field": "geo.src" }},
      "agent_dc": { "cardinality": { "field": "agent.keyword" }},
      "geo.dest_dc": { "cardinality": { "field": "geo.dest" }},
      "responses.total": { "value_count": { "field": "timestamp" }},
      "success" : { 
         "filter": {
            "term": { "response" : "200"}}
        },
      "error404" : {
         "filter": {
            "term": { "response" : "404"}}
        },
      "error503" : {
         "filter": {
            "term": { "response" : "503"}}
        },
      "timestamp.min": { "min": { "field": "timestamp" }},
      "timestamp.max": { "max": { "field": "timestamp" }},
      "timestamp.duration_ms": { 
        "bucket_script": {
          "buckets_path": {
            "min_time": "timestamp.min.value",
            "max_time": "timestamp.max.value"
          },
          "script": "(params.max_time - params.min_time)"
        }
      }
    }
  }
}

The destination index for the transform.

Configures the transform to run continuously. It uses the timestamp field to synchronize the source and destination indices. The worst case ingestion delay is 60 seconds.

The data is grouped by the clientip field.

Filter aggregation that counts the occurrences of successful (200) responses in the response field. The following two aggregations (error404 and error503) count the error responses by error codes.

This bucket_script calculates the duration of the clientip access based on the results of the aggregation.

After you create the transform, you must start it:

POST _transform/suspicious_client_ips/_start

Shortly thereafter, the first results should be available in the destination index:

GET sample_weblogs_by_clientip/_search

The search result shows you data like this for each client IP:

    "hits" : [
      {
        "_index" : "sample_weblogs_by_clientip",
        "_id" : "MOeHH_cUL5urmartKj-b5UQAAAAAAAAA",
        "_score" : 1.0,
        "_source" : {
          "geo" : {
            "src_dc" : 2.0,
            "dest_dc" : 2.0
          },
          "success" : 2,
          "error404" : 0,
          "error503" : 0,
          "clientip" : "0.72.176.46",
          "agent_dc" : 2.0,
          "bytes_sum" : 4422.0,
          "responses" : {
            "total" : 2.0
          },
          "url_dc" : 2.0,
          "timestamp" : {
            "duration_ms" : 5.2191698E8,
            "min" : "2020-03-16T07:51:57.333Z",
            "max" : "2020-03-22T08:50:34.313Z"
          }
        }
      }
    ]

Like other Kibana sample data sets, the web log sample dataset contains timestamps relative to when you installed it, including timestamps in the future. The continuous transform will pick up the data points once they are in the past. If you installed the web log sample dataset some time ago, you can uninstall and reinstall it and the timestamps will change.

This transform makes it easier to answer questions such as:

  • Which client IPs are transferring the most amounts of data?
  • Which client IPs are interacting with a high number of different URLs?
  • Which client IPs have high error rates?
  • Which client IPs are interacting with a high number of destination countries?

Finding the last log event for each IP addressedit

This example uses the web log sample data set to find the last log from an IP address. Let’s use the latest type of transform in continuous mode. It copies the most recent document for each unique key from the source index to the destination index and updates the destination index as new data comes into the source index.

Pick the clientip field as the unique key; the data is grouped by this field. Select timestamp as the date field that sorts the data chronologically. For continuous mode, specify a date field that is used to identify new documents, and an interval between checks for changes in the source index.

This transform creates the destination index that contains the latest login date for each client IP. As the transform runs in continuous mode, the destination index will be updated as new data that comes into the source index.

PUT _transform/last-log-from-clientip
{
  "source": {
    "index": [
      "kibana_sample_data_logs"
    ]
  },
  "latest": {
    "unique_key": [ 
      "clientip"
    ],
    "sort": "timestamp" 
  },
  "frequency": "1m", 
  "dest": {
    "index": "last-log-from-clientip"
  },
  "sync": { 
    "time": {
      "field": "timestamp",
      "delay": "60s"
    }
  },
  "settings": {
    "max_page_search_size": 500
  }
}

Specifies the field for grouping the data.

Specifies the date field that is used for sorting the data.

Sets the interval for the transform to check for changes in the source index.

Contains the time field and delay settings used to synchronize the source and destination indices.

After you create the transform, start it:

POST _transform/last-log-from-clientip/_start

After the transform processes the data, search the destination index:

GET last-log-from-clientip/_search

The search result shows you data like this for each client IP:

{
  "_index" : "last-log-from-clientip",
  "_id" : "MOeHH_cUL5urmartKj-b5UQAAAAAAAAA",
  "_score" : 1.0,
  "_source" : {
    "referer" : "http://twitter.com/error/don-lind",
    "request" : "/elasticsearch",
    "agent" : "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET CLR 1.1.4322)",
    "extension" : "",
    "memory" : null,
    "ip" : "0.72.176.46",
    "index" : "kibana_sample_data_logs",
    "message" : "0.72.176.46 - - [2018-09-18T06:31:00.572Z] \"GET /elasticsearch HTTP/1.1\" 200 7065 \"-\" \"Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET CLR 1.1.4322)\"",
    "url" : "https://www.elastic.co/downloads/elasticsearch",
    "tags" : [
      "success",
      "info"
    ],
    "geo" : {
      "srcdest" : "IN:PH",
      "src" : "IN",
      "coordinates" : {
        "lon" : -124.1127917,
        "lat" : 40.80338889
      },
      "dest" : "PH"
    },
    "utc_time" : "2021-05-04T06:31:00.572Z",
    "bytes" : 7065,
    "machine" : {
      "os" : "ios",
      "ram" : 12884901888
    },
    "response" : 200,
    "clientip" : "0.72.176.46",
    "host" : "www.elastic.co",
    "event" : {
      "dataset" : "sample_web_logs"
    },
    "phpmemory" : null,
    "timestamp" : "2021-05-04T06:31:00.572Z"
  }
}

This transform makes it easier to answer questions such as:

  • What was the most recent log event associated with a specific IP address?
« Tutorial: Transforming the eCommerce sample data Painless examples for transforms »