The application has users and groups. Based on the groups the users are in, will have access to locations. IoT devices may/may not be linked to locations but will still be streaming values.
There is a table displayed on the front-end for which data is coming from DynamoDB. We will be updating the last measurement value for each of those devices in DynamoDB based on measurement values streaming in from IoT devices. There are going to be millions of these devices.
I am considering using the approach mentioned here https://aws.amazon.com/blogs/mobile/iot-with-aws-appsync/
Is there a simpler way to do it? We are okay with updating values only once a second.
I am wondering how many topics we will end up with? Should it be a topic per device? or a topic per location? that the client will subscribe to.
Does the client subscribe to the topic only when they land on the screen and unsubscribe when they land away?
The "last value from IoT devices" is to be updated only in one column of a data table in UI while others remain the same. It should still work, Is that right?
I also see some solutions using Kinesis firehose and analytics. Not sure if that's relevant for this case. Any simpler approaches to do this?
与恶龙缠斗过久,自身亦成为恶龙;凝视深渊过久,深渊将回以凝视…