Amazon DynamoDB Batch Put

This nodes writes KNIME table rows as items into DynamoDB. The KNIME table's columns are the attributes of the written DynamoDB item, which means that the table must contain matching columns for the DynamoDB table's hash and range key attributes. If the table does not have a range key, only a column with the name and type of the hash key must be present.

The command sent to DynamoDB by this node is BatchWriteItem, with a single batch writing at most 25 items, so inserting a large amount of data may take a while. The individual write operations on item level are atomic, but the operation is not atomic on batch or even node execution level. That means if the node fails in-between, some of the items may have been written and others not.

A single batch can write at most 16 MB of data and a single item can be as large as 400 KB.

If DynamoDB throttles the requests due to insufficient provisioned capacity units, unprocessed items are included in the next request. Additionally this node employs exponential backoff if throttling occurs. Each time a request cannot be completed fully, the next request is sent after a waiting time of 2^nRetry * 100 milliseconds. Once a request can be executed completely, the retry counter is reset. However, if no item can be deleted, the node fails and you have to increase the provisioned write capacity units on the table.

More information can be found in the DynamoDB documentation.

Options

Region
The region the table is in.
Table Name
The table to access.
Custom Endpoint (*)
A custom endpoint if the default AWS endpoint should not be used, e.g. for DynamoDB Local.
Batch Size
The number of items to delete in a single batch (max. 25, min. 1).
Publish consumed capacity units as flow variable
If checked, the total capacity units used by the operation are published as a flow variable named "batchPutConsumedCapacity".

Input Ports

Icon
Credentials for an AWS account
Icon
KNIME data table

Output Ports

Icon
Credentials for an AWS account

Views

This node has no views

Workflows

Links

Developers

You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.