Using AWS CloudFormation to Set Up Neptune-to-Neptune Replication with the Streams Consumer Application

VPC   –   Provide the name of the VPC where the polling Lambda function will run.

SubnetIDs   –   The subnets to which a network interface is established. Add subnets corresponding to your Neptune cluster.

SecurityGroupIds   –   Provide the IDs of security groups that grant write inbound access to your source Neptune DB cluster.

RouteTableIds   –   This is needed to create an Amazon DynamoDB endpoint in your Neptune VPC, if you do not already have one. You must provide a comma-separated list of route table IDs associated with the subnets.

CreateDDBVPCEndPoint   –   A Boolean value that defaults to true, indicating whether or not it is necessary to create a Dynamo DB VPC endpoint. You only need to change it to false if you have already created a DynamoDB endpoint in your VPC.

CreateMonitoringEndPoint   –   A Boolean value that defaults to true, indicating whether or not it is necessary to create a monitoring VPC endpoint.. You only need to change it to false if you have already created a monitoring endpoint in your VPC.

ApplicationName   –   You can generally leave this set to the default (NeptuneStream). If you use a different name, it must be unique.

LambdaMemorySize   –   Used to set the memory size available to the Lambda poller function. The default value is 2,048 megabytes.

LambdaRuntime   –   The language used in the Lambda function that retrieves items from the Neptune stream. You can set this either to python3.9 or to java8.

LambdaS3Bucket   –   The Amazon S3 bucket that contains Lambda code artifacts. Leave this blank unless you are using a custom Lambda polling function that loads from a different Amazon S3 bucket.

LambdaS3Key   –   The Amazon S3 key that corresponds to your Lambda code artifacts. Leave this blank unless you are using a custom Lambda polling function.

LambdaLoggingLevel   –   In general, leave this set to the default value, which is INFO.

ManagedPolicies   –   Lists the managed policies to use for execution of your Lambda function. In general, leave this blank unless you are using a custom Lambda polling function.

StreamRecordsHandler   –   In general, leave this blank unless you are using a custom handler for the records in Neptune streams.

StreamRecordsBatchSize   –   The maximum number of records to be fetched from stream. You can use this parameter to tune performance. The default (5000) is a good place to start. The maximum allowable is 10,000. The higher the number, the fewer network calls are needed to read records from the stream, but the more memory is required to process the records. Lower values of this parameter result in lower throughput.

MaxPollingWaitTime   –   The maximum wait time between two polls (in seconds). Determines how frequently the Lambda poller is invoked to poll the Neptune streams. Set this value to 0 for continuous polling. The maximum value is 3,600 seconds (1 hour). The default value (60 seconds) is a good place to start, depending on how fast your graph data changes.

MaxPollingInterval   –   The maximum continuous polling period (in seconds). Use this to set a timeout for the Lambda polling function. The value should be in the range between 5 seconds and 900 seconds. The default value (600 seconds) is a good place to start.

StepFunctionFallbackPeriod   –   The number of units of step-function-fallback-period to wait for the poller, after which the step function is called through Amazon CloudWatch Events to recover from a failure. The default (5 minutes) is a good place to start.

StepFunctionFallbackPeriodUnit   –   The time units used to measure the preceding StepFunctionFallbackPeriodUnit (minutes, hours, or days). The default (minutes) is generally sufficient.

NeptuneStreamEndpoint   –   (Required) The endpoint of the Neptune source stream. This takes one of two forms:

Neptune Query Engine   –   Choose Gremlin, openCypher, or SPARQL.

IAMAuthEnabledOnSourceStream   –   If your Neptune DB cluster is using IAM authentication, set this parameter to true.

StreamDBClusterResourceId   –   If your Neptune DB cluster is using IAM authentication, set this parameter to the cluster resource ID. The resource ID is not the same as the cluster ID. Instead, it takes the form: cluster- followed by 28 alpha-numeric characters. It can be found under Cluster Details in the Neptune console.

TargetNeptuneClusterEndpoint   –   The cluster endpoint (hostname only) of the target backup cluster.

Note that if you specify TargetNeptuneClusterEndpoint, you cannot also specify TargetSPARQLUpdateEndpoint.

TargetNeptuneClusterPort   –   The port number for the target cluster.

Note that if you specify TargetSPARQLUpdateEndpoint, the setting for TargetNeptuneClusterPort is ignored.

IAMAuthEnabledOnTargetCluster   –   Set to true if IAM authentication is to be enabled on the target cluster.

TargetAWSRegion   –   The target backup cluster's AWS region, such as us-east-1). You must provide this parameter only when the AWS region of the target backup cluster is different from the region of the Neptune source cluster, as in the case of cross-region replication. If the source and target regions are the same, this parameter is optional.

Note that if the TargetAWSRegion value is not a valid AWS region that Neptune supports, the process fails.

TargetNeptuneDBClusterResourceId   –   Optional: this is only needed when IAM authentication is enabled on the target DB cluster. Set to the resource ID of the target cluster.

SPARQLTripleOnlyMode   –   Boolean flag that determines whether triple-only mode is enabled. In triple-only mode, there is no named-graph replication. The default value is false.

TargetSPARQLUpdateEndpoint   –   URL of the target endpoint for SPARQL update, such as https://abc.com/xyz. This endpoint can be any SPARQL store that supports quad or triples.

Note that if you specify TargetSPARQLUpdateEndpoint, you cannot also specify TargetNeptuneClusterEndpoint, and the setting of TargetNeptuneClusterPort is ignored.

BlockSparqlReplicationOnBlankNode   –   Boolean flag which, if set to true, stops replication for BlankNode in SPARQL (RDF) data. The default value is false.

Required to create Cloud watch Alarm   –   Set this to true if you want to create a CloudWatch alarm for the new stack.

SNS Topic ARN for Cloudwatch Alarm Notifications   –   The SNS topic ARN where CloudWatch alarm notifications should be sent (only needed if alarms are enabled).

Email for Alarm Notifications   –   The email address to which alarm notifications should be sent (only needed if alarms are enabled).