1. Administering Oracle Blockchain Platform on Oracle Cloud Infrastructure (Gen 2)
  2. Manage the Life Cycle of an Instance
  3. Scale Your Instance

Scale Your Instance

You can scale your overall instance OCPU capacity or the components of any instance.

Scaling the OCPU Capacity of an Enterprise Instance

If you've selected an Enterprise edition when provisioning your instance, you can scale its OCPU capacity up or down. When you scale OCPU capacity, the platform shape changes to Custom in the Instance Details view. The total OCPU capacity represents the total number of OCPUs that are equally allocated among all the VMs in the Oracle Blockchain Platform cluster.

To change your OCPU capacity:
  1. In the Oracle Cloud Infrastructure console, under Developer Services, select Blockchain Platform.
  2. Select the correct compartment, and click the Enterprise instance that you want to modify.
  3. In the Instance Details view, click Scale, and then from the Select Scale Object drop-down list select Platform.
  4. You are prompted to select the number of OCPUs that you want provisioned for this instance.

    Note that the resizing process will take several minutes and will result in downtime of your Oracle Blockchain Platform instance.

Your instance is now resized to its new OCPU capacity and is listed as a customized edition.

Scaling Components of Your Instance

You can scale components of your instance by selecting it in the Oracle Cloud Infrastructure console and clicking Scale.

Peers
  • You can change the OCPUs for the peer nodes. The total value of the OCPUs must not exceed the size of the instance edition that you selected when scaling your instance. After you select the OCPUs for a peer, you can decrease the OCPU allocation if required.
  • You can add a new peer by clicking Add Row and selecting the alias, OCPUs, role, and availability domain or fault domain for that peer.

Note:

When scaling peers, the OCPUs field is a logical representation of processing power, which is used to determine whether a newly scaled peer will be placed on an existing VM or a new VM. It does not map to the actual number of OCPU cores enabled on the virtual machine (VM).

You can see the availability domains or fault domains for the existing peers on the Scale Blockchain Platform window. Each Oracle Blockchain Platform instance is deployed across three availability domains or fault domains.

You can use the following REST API call to get more information about the logical OCPU consumption for each VM: Get Blockchain Platform Information.

The peers array in the BlockchainPlatformComponentDetails object shows information for each peer, including the host where the peer is located and the ocpuAllocationNumber, which indicates the number of OCPUs allocated to the peer.

The hostOcpuUtilizationInfo array shows the ocpuCapacityNumber and ocpuUtilizationNumber values for each VM (host). To determine the remaining OCPU capacity that can be allocated on a VM, subtract the ocpuUtilizationNumber value from the ocpuCapacityNumber value.

When you add a peer to a particular availability domain or fault domain, if the specified OCPUs value is greater than the remaining OCPU capacity on the VM in that domain, then a new VM is allocated. Otherwise, the existing VM is used to host the new peer. For domains that already have multiple VMs, if at least one VM has enough OCPU capacity, the peer is hosted there. Otherwise, a new VM is allocated.

Ordering Service
  • You can add an additional orderer node if you have selected to provision an Enterprise Edition instance.
REST Proxy
  • You can change the number of cluster replicas for your REST proxy.
Certificate Authority:
  • You can change the number of cluster replicas for your certificate authority.
Storage
  • Storage scaling is for Enterprise shapes only.
  • You can add storage in 1TB increments. Metering of storage is also in 1TB increments. You can't decrease storage once allocated.
  • The initial block size is 50GB for each manager VM in the cluster, and is increased in increments of 100GB.
  • When storage is scaled up on an instance, the block volume on each VM in the cluster will be increased by the same size. All VMs in a cluster will have block storage volumes of the same size at all times.
  • When new VMs are added as part of scaling peers and OSNs, every new VM will be brought to parity with existing VMs in the cluster by adding the same number and size of storage volumes as present in an existing VM. This additional storage will be taken from your platform storage pool (storage which has been purchased but not yet allocated to a VM). In situations when the platform storage pool does not have sufficient capacity to create the same number and size of block volumes for the new VM:
    • If the additional storage required is less than 0.5TB of the total purchased (metered) storage, we will add additional storage to enable creation of required volumes, and you will not be charged for additional <0.5TB of storage.
    • If the additional storage required is more than 0.5TB of the total purchased (metered) storage, you will be charged for additional storage units as applicable.