Batch Bytes

This topic describes how to use the function in the .

Description

The Batch Bytes function batches incoming byte arrays by size, count, or milliseconds and outputs a single byte array concatenated by a user defined separator.

There are two functions for batching records: Batch Bytes and Batch Records. Use Batch Bytes when you need to serialize your data before batching. Use Batch Records when you want to do serialization (if any) after batching. Batch Bytes concatenates the batched byte array payloads delimited by an optional user-defined separator. This means that Batch Bytes is best used for row-based style batched payloads, such as CSV or JSON streaming (concatenated JSON records), but it will not be useful for bulk or columnar formats.

Function Input/Output Schema

Function Input

byte[]

This function takes in byte arrays to be batched.

Function Output

byte[]

This function outputs a byte array that is a concatenation of all byte[] payloads.

Syntax

The required syntax is in bold.

batch_bytes

bytes: <byte array>

separator: <string>

size: <string> B | KB | MB

num_events: <long>

millis: <long>

Required arguments

bytes

Syntax: byte[]

Description: The byte array payload to be batched.

Optional arguments

separator

Syntax: <string>

Description: A delimiter that separates the byte payloads.

Example: \\n

size

Syntax: <string> B | KB | MB

Description: The maximum size, in bytes, of the emitted batched byte[]. The size of your emitted batched bytes cannot exceed 100 MB.

Default: 10MB

Example: 1024B

num_events

Syntax: <long>

Description: The maximum number of payloads per batch.

Default: 100,000,000

Example: 2000

millis

Syntax: <long>

Description: The interval, in milliseconds, at which to send batched records. Cannot exceed 8000 milliseconds (8 seconds).

Default: 2000 milliseconds (2 seconds).

Example: 2000

Usage

The following is an example of batched data. Assume that your data looks something like the following snippet, and you've configured your function with the arguments shown in the SPL2 example.

{"event": "my data 1", "index": "syslog", "sourcetype": "syslog"}
{"event": "my data 2", "index": "secindex", "sourcetype": "security"}
{"event": "my data 3", "index": "proxy", "sourcetype": "squid"}
....

When the batch trigger fires, Batch Bytes outputs a single record with a byte array.

Record {
   "bytes": byte[]
}

The outputted byte array is the following string as UTF-8 bytes.

{"event": "my data 1", "index": "syslog", "sourcetype": "syslog"}{"event": "my data 2", "index": "secindex", "sourcetype": "security"}{"event": "my data 3", "index": "proxy", "sourcetype": "squid"}

Example

An example of a common use case follows. These examples assume that you have added the function to your pipeline.

SPL2 Example: Batch Splunk JSON events and send the payload once it reaches 5MB or 2 minutes have passed

This example assumes that you are in the SPL View.

Suppose you have Splunk JSON events outputted by the to_splunk_json function. To batch bytes of those Splunk JSON events and only send payloads once they reach 5MB or 2 minutes have passed:

... | to_splunk_json index=cast(map_get(attributes, "index"), "string")  | batch_bytes bytes=to_bytes(json) size="5MB" millis="120000" |...;