Google Cloud Bigtable Memorystore

This code shows how to use Cloud Bigtable with a caching solution hosted on Memorystore.

Setup

1. Create a Bigtable instance

2. Set your variables

   BIGTABLE_PROJECT=YOUR-PROJECT-ID OR $GOOGLE_CLOUD_PROJECT
   INSTANCE_ID=YOUR-INSTANCE-ID
   TABLE_ID=mobile-time-series # Example table for sample  
   

3. Create a table with one row

   echo project = $BIGTABLE_PROJECT > ~/.cbtrc
   echo instance = $INSTANCE_ID >> ~/.cbtrc
   
   cbt createtable $TABLE_ID "families=stats_summary" 
   cbt set mobile-time-series phone#4c410523#20190501 stats_summary:os_build=PQ2A.190405.003 stats_summary:os_name=android
   

Memcached

1. Set up a Memcached instance locally or hosted on Memorystore.

2. Get the Memcached host IP and set that as a variable.

   MEMCACHED_DISCOVERY_ENDPOINT="0.0.0.0"
   

3. Run the code. If you are trying to connect to Memcached on Memorystore, you will need to run this code within the same VPC network to connect. The easiest way to do this is by creating a Compute VM on the same network, SSHing to that and running the code from there.

   mvn compile exec:java -Dexec.mainClass=Memcached \
   -DbigtableProjectId=$PROJECT_ID \
   -DbigtableInstanceId=$INSTANCE_ID \
   -DbigtableTableId=$TABLE_ID \
   -DmemcachedDiscoveryEndpoint=$MEMCACHED_DISCOVERY_ENDPOINT
   

   The first time you run the program, it will fetch the data from Bigtable. If you run it again (within the cache limit), it will fetch the data from Memcached.