I am using aws-sdk to connect with DynamoDb and I have got into a scenario where I got one dynamodb table with different partition/hash-key and I have to scan and filter to get the results. Scanning the entire table would be a costly operation . Is there a way to scan only a certain partion/haskey of a table ?
1
-
If I understood correctly, you need to look into queries. They allow you to get data based on primary key (partition/hash)kaskelotti– kaskelotti2019-03-25 14:05:01 +00:00Commented Mar 25, 2019 at 14:05
Add a comment
|
1 Answer
You have to use Dynamo DB Query. You can query any table or secondary index that has a composite primary key (a partition key and a sort key).
In my opinion you shouldn't use scan, because it very costly and slow.
You didn't write what is the program language, but here is some example to query:
https://docs.aws.amazon.com/AWSJavaScriptSDK/latest/AWS/DynamoDB.html#query-property
https://docs.aws.amazon.com/amazondynamodb/latest/APIReference/API_Query.html
About indices:
UPDATE #1:
Maybe that will help:
Add a new column to your table. The values will be static. (For example: Col name: const_value Values: const)
Create a new secondary index to your table.
- 'partition key': 'const_value'
- 'sort key': the column what you want to filter
You can use Query.
2 Comments
Ashwin Dora
Yes i have already used DynamoDB Query in a way to make it work but the problem with that is every item getting inserted to my table has unique partition/sort key and that will not work because based on a filter criteria I have to get the results.
Ashwin Dora
I like the idea but lets say that I create a GSI with its PK/HK as a constant and SK is a string and it could be only like apples or oranges will it work ? I am not sure if it will work because I agree that we can have 1 bucket/PK/HK as constant but the SK needs to be unqiue in that case isnt it ? Or do you think SK can repeat ?