crux-aws

crux atop S3 and DynamoDB.

Experimental.

Quick Start:

For a full AWS experience:

(def node (crux.api/start-node {:crux.node/topology crux.aws/topology
                                :crux.dynamodb/table-name "your-dynamodb-tx-table"
                                :crux.s3/bucket "your-s3-doc-bucket"
                                :crux.kv.hitchhiker-tree/konserve crux.kv.hitchhiker-tree.konserve.ddb-s3/ddb-s3-backend
                                :crux.kv.hitchhiker-tree.konserve.ddb-s3/bucket "your-dynamodb-kv-table"
                                :crux.kv.hitchhiker-tree.konserve.ddb-s3/table "your-s3-kv-bucket"
                                :crux.kv.hitchhiker-tree.konserve.ddb-s3/region "us-west-2"}))

NOTE the hitchhiker-tree KV store with the konserve-ddb-s3 backend is currently broken; use a different KV store for now.

Subprojects you should look at as well:

• crux-dynamodb -- TxLog on DynamoDB.
• crux-ddb-s3 -- TxLog spread between DynamoDB and S3.
• crux-hitchhiker-tree -- KV store on hitchhiker-tree.

TODO

The KV store needs to handle concurrent writes better than it does.

The KV store should take updates from the remote store periodically (or on changes to dynamodb) so it keeps in sync.

The transaction log could wait for changes triggered