What is Oracle Data Hub Cloud Service (DHCS)?
The Oracle Data Hub Cloud Service (DHCS) is a self-managed cloud service enabling customers to easily provision, administer and monitor popular open source database clusters within the Oracle Cloud platform.
What can I do with the DHCS?
Currently, you can use the DHCS Console/API/CLI to easily provision the Apache Cassandra database clusters within the Oracle Cloud Infrastructure Classic (OCI-Classic) platform. You can use this cluster as a data store for your big-data, cloud-native applications or to persistently store the messages from the Oracle Event Hub Cloud Service.
When should I use the DHCS?
You should use the DHCS when you want a consistent interface to provision, administer and monitor popular open source database clusters within the Oracle Cloud platform. For example, if you either manage the Apache Cassandra cluster in an on-premise environment or build an application that requires a linearly-scalable, distributed NoSQL data-store such as Apache Cassandra, then you should use the Oracle Data Hub Cloud Service so that you can easily create, monitor and administer such Apache Cassandra clusters within the Oracle cloud platform.
How can I get started with DHCS?
If you have an Oracle Cloud account with Universal Credits, then you can easily get started by creating the Apache Cassandra cluster using the Oracle Data Hub Cloud Service console or CLI. Once you have created the clusters, then you can use the database native tools (such as 'cqlsh') to connect to the Apache Cassandra database cluster and perform database specific tasks. The DHCS console/API/CLI interface allows you to easily manage, administer and monitor your clusters. For more information, check out our Getting Started product documentation.
If you do not have an Oracle Cloud account with Universal Credits, then you can easily sign up for the free $300 credits. For more information, check out our Try-it promotion.
What are the prerequisites to create a cluster within the DHCS in OCI-Classic region?
You will need to configure replication policy within the Object Storage Classic region.
How can I successfully create a cluster in DHCS?
You can create the cluster by choosing the number of nodes in the cluster, the compute shape for the nodes in the cluster, and the block volume size for storing the database within every node in the cluster. Optionally, you can also provide the object storage URL and access information so that the DHCS can regularly backup your database to the object storage.
What are the available compute shapes for the clusters?
When you create the cluster on the OCI-Classic platform, you can choose between the standard or High Memory virtual machine shapes.
What is the default cluster size?
The default cluster size is 3 nodes within the cluster wherein each node is a 1 OCPU.
How can I access the nodes in the cluster?
You will be able to "ssh" to all the nodes in the cluster.
Can I change the configuration files within the node to meet my requirement?
Yes, you have full access to all the nodes in the cluster. You can login to the nodes using the secure shell and modify the database configuration files as appropriate. If you make any configuration changes outside the database configuration files, then Oracle cannot support you.
Can I create clusters across multiple data centers or regions?
Currently, you can create clusters within a data center in OCI-Classic platform..
Can I dynamically expand or shrink the cluster size?
Yes. You can use the console/API to easily add a node (scale-out) or remove a node (scale-in) from the cluster to meet your throughput requirements.
How can I add or remove a node in the cluster?
You can either use our CLI/API/Console to either add or remove a node to the cluster. The service will replicate the data as part of this scale-out or scale-in operation to avoid any data loss.
Where is the data actually stored?
The DHCS configures the nodes in the cluster to write the database data and the commit log to the block volumes attached to the compute nodes.
What are the available cloud regions where I can create the DHCS clusters?
You can use the DHCS to create the database clusters in the following cloud regions:
- us-chicago-1 (OCI Classic-North America)
- us-ashburn-1 (OCI Classic-North America)
- uk-london-1 (OCI Classic-Europe)
- eu-frankfurt-1 (OCI Classic-Europe)
Can I use DHCS to create the database clusters at Cloud@Customer?
Currently, this is in our roadmap. Please reach out to us so that we can understand your use case.
How can my application running in ACCS connect to the DHCS cluster?
Within your application, you can connect either by using the DHCS node IP-address and port 9042 combination or by using the ACCS service binding feature. When you use the latter, you can simply connect to the DHCS cluster instance using an ACCS provided environment variable. This capability allows you to easily update your DHCS cluster instance without causing any impact to your ACCS→DHCS connectivity. For more information, reference the ACCS product documentation-Managing Service Bindings.
Data Hub Cloud Service for Apache Cassandra
How can I administer the Apache Cassandra database cluster created within DHCS?
You can use the DHCS Console/API/CLI to administer your database clusters such as adding storage to the nodes in the cluster, scaling-up (adding more OCPUs to the nodes in the cluster), scaling-out (adding a node to the cluster) and scaling-in (removing a node from the cluster).
How can I monitor the Apache Cassandra cluster?
You can easily monitor the cluster either by using the DHCS REST API or the Service Console. The service console is accessible via https://<public IP address for any of the nodes in the cluster>:8001
How does the billing work?
You only pay for what you use on an hourly basis. There is no separate minimum or one-time setup charges. Your hourly bill will be based on:
- Used OCPUs: You will pay depending on the compute instance shape and the number of OCPUs used across your clusters.
- Allocated Block Volumes: You will pay the total allocated block volumes across your clusters.
- Object Storage Volumes: You will be pay the actual GiB of data used to backup your database clusters. You can choose whether you want to backup to the Object Store. If you don't choose to backup, then you will not incur any fees here.
Let us say, you sign up for the Oracle Cloud pay-as-you-go without any commitment and use the Data Hub Cloud Service to run a 3-node (with 1-OCPU each) Apache Cassandra Database cluster for a month with 400 GB allocated block volume storage per node and a 500 GB of backup to Object Storage. Your monthly cost will be $462.
- Compute (OCPUs): $0.1815 /OCPU /Hour * 3 * 730 = $397.485
- Allocated Block Volumes: $0.0425 /GiB/Month * 400 * 3 = $51
- Object Storage Volume: $0.0255 /GiB/Month * 500 = $12.75