DynamoDB Basic APIs

Links: 102 AWS DVA Index

---

Writing Data

• PutItem

  • Creates a new item or fully replace an old item (same Primary Key)
  • Consumes WCUs

• UpdateItem

  • Edits an existing item's attributes or adds a new item if it doesn't exist
    • Like Upsert.
  • Can be used to implement Atomic Counters a numeric attribute that's unconditionally incremented.
    • With an atomic counter, the numeric value will increment each time you call UpdateItem.

• ConditionalWrites

  • By default, the DynamoDB write operations (PutItem, UpdateItem, DeleteItem) are unconditional.
  • Accept a write/update/delete (PutItem/UpdateItem/DeleteItem) only if conditions are met, otherwise returns an error.
  • Helps with concurrent access to items
  • Prevents overwriting.

APIs for writing data: PutItem, UpdateItem and DeleteItem.

Reading Data

• GetItem
  • Read based on Primary key
  • Gets only a single item.
  • Primary Key can be HASH or HASH+RANGE
  • Eventually Consistent Read (default)
  • Option to use Strongly Consistent Reads (more RCU, might take longer, higher latency)
    • For consistent reads Set the ConsistentRead parameter to true when calling GetItem.
  • ProjectionExpression (--projection-expression): can be specified to retrieve only certain attributes (columns)
    • A projection expression is a string that identifies the attributes you want.
    • To retrieve a single attribute, specify its name.
    • For multiple attributes, the names must be comma-separated.

Query

• Query returns items (rows) based on:
  • KeyConditionExpression: (--query)
    • Partition Key value (must be = operator): required
    • Sort Key value (=, <, <=, >, >=, Between, Begins with): Optionally, you can provide a sort key attribute and use a comparison operator to refine the search results.
  • FilterExpression (--filter-expression)
    • Additional filtering after the Query operation (before data returned to you)
    • Use only with non-key attributes (does not allow HASH or RANGE attributes)
• Returns:
  • The number of items specified in Limit
  • Or up to 1MB of data
• To get more data we need to use pagination.
• Can query table, a Local Secondary Index, or Global Secondary Index.

Scan

GetItem was for one item, query was for specific partition key and scan is to read out an entire table and then filter our data. It is inefficient.

Filtering has to be done on the client side

• Returns up to 1 MB of data use pagination to keep on reading
• Consumes a lot of RCU
• Limit impact using Limit or reduce the size of the result and pause
• For faster performance, use Parallel Scan
  • Multiple workers scan multiple data segments at the same time
  • Increases the throughput and RCU consumed
  • Limit the impact of parallel scans just like you would for Scans
• Can use ProjectionExpression & FilterExpression (no changes to RCU)

How to scan the whole table in the shortest possible time whilst ensuring the normal workload is NOT affected.

Use the Parallel Scan API operation and limit the rate.

Perform a rate-limited parallel scan operation.

Deleting Data

• DeleteItem
  • Delete an individual item
  • Ability to perform a conditional delete
• DeleteTable
  • Delete a whole table and all its items
  • Much quicker deletion than calling DeleteItem on all items

Batch Operations

• Allows you to save in latency by reducing the number of API calls
  • Reduces network overhead of multiple transactions.
• Operations are done in parallel for better efficiency.
• Part of a batch can fail; in which case we need to try again for the failed items
• BatchWriteItem
  • Up to 25 PutItem and/or DeleteItem in one call
  • Up to 16 MB of data written, up to 400 KB of data per item
  • Can't update items (use UpdateItem)

We DON'T have batch operations for updating items (UpdateItem).

• BatchGetItem
  • Return items from one or more tables
  • Up to 100 items, up to 16 MB of data
  • Items are retrieved in parallel to minimise latency

An application needs to read up to 100 items at a time from an Amazon DynamoDB. Each item is up to 100 KB in size and all attributes must be retrieved. What is the BEST way to minimise latency?

Use BatchGetItem.

Returning WCU consumed

• For operations PutItem, UpdateItem and DeleteItem.
• To return the number of WCU consumed by any of these operations, set the ReturnConsumedCapacity parameter to one of the following:
  • TOTAL - returns the total number of write capacity units consumed.
  • INDEXES - returns the total number of WCU consumed, with subtotals for the table and any secondary indexes that were affected by the operation.
    • If you see the word indexes in the question then go for INDEXES.
  • NONE - no write capacity details are returned. (This is the default.)

---

Last updated: 2023-02-20