Application resource demand changes over time with traffic, business activity, scheduled processing, and other workload patterns. Customers typically configure compute and memory capacity to support periods of peak demand. When demand is lower, the same workloads can remain overprovisioned because their configured capacity stays the same while actual resource use declines.
Rightsizing can reduce unused capacity, but it reflects workload needs at a point in time and may not account for continuing changes in demand. Reducing configured capacity too far can also limit the capacity headroom available when resource demand returns.
Today, we are excited to announce the general availability of Dynamic Resource Virtual Machines (DRVM) for Oracle Cloud Infrastructure Compute. DRVM gives customers a more efficient and economical way to maintain the OCPU and memory capacity their applications require while OCI dynamically manages the physical resources backing each VM.
DRVM is available for eligible VM.Standard.E6.Flex configurations in select OCI regions, with additional OCI Compute shapes and regions planned in the following months.
The preceding is intended to outline our general product direction, It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code or functionality, and should not be relied upon in making purchasing decisions. The development, release, timing and pricing of any features or functionality described for Oracle’s products may change and remains at the sole discretion of Oracle Corporation.
What is a Dynamic Resource VM?
DRVM is a resource-management option for eligible VMs. Customers continue to select the number of OCPUs and the amount of memory required by their applications. Those values remain the configured capacity of the VM.
For example, a DRVM configured with 8 OCPUs and 64 GB of memory can use up to all 8 OCPUs and 64 GB of memory. OCI dynamically manages how the underlying physical CPU and memory are allocated as workload demand changes.
Compared to OCI Burstable instances, DRVM instances do not require a resource utilization baseline, and there is no requirement to accumulate burst credits before using the configured capacity. OCI Compute dynamically manages the underlying allocation of the CPU and Memory resources to support each DRVM instance’s workload demand.
Why use DRVM?
DRVM helps reduce the cost of maintaining capacity for workloads whose resource needs can change over time. Customers can retain capacity for higher-demand periods without paying the full Standard VM shape rate during the entire life of the instance. Key benefits include:
- Lower OCPU and memory usage costs
- Access to the full configured OCPU and memory capacity when the workload requires it
- Uses the existing Compute instance shape configuration and lifecycle APIs for supported DRVM operations.
While DRVM does not replace good capacity planning or rightsizing, it provides an additional resource optimization option for applications that need capacity headroom but do not continuously use all of it.
When to use DRVM
DRVM is best suited for general-purpose workloads with variable, cyclical, or intermittent OCPU and memory demand. Good candidates can include websites, applications, and API endpoints with changing traffic patterns, stateless services, microservices, development and test environments, CI/CD workers, and batch or ETL workloads with alternating periods of higher and lower activity. Applications distributed across multiple VMs can also be a good fit when their resource peaks do not always occur at the same time.
Standard VMs should remain the preferred choice for workloads with sustained or highly consistent OCPU and memory demand. Standard is also the better choice when the predictable and consistent resource behavior is more important than lower costs. This can include latency-sensitive applications, database workloads with sustained resource demand, real-time processing, or workloads with tightly synchronized demand across many instances. Overall, DRVM provides a great combination of price and performance though it is not a universal replacement for Standard VMs.
DRVM economics
DRVM instances receive a 10% discount on their OCPU and memory usage. For example, an E6 Flex VM configured with 8 OCPUs and 64 GB of memory using dynamic resource management receives the same configured OCPU and memory capacity at a 10% lower cost than the equivalent Standard E6 Flex configuration. Storage, networking, operating-system licenses, and other separately billed services retain their applicable rates.
How DRVM works
DRVM combines several OCI Compute capabilities to manage physical resources while preserving the configured VM experience.
- Dynamic OCPU scheduling: Virtual CPU work is scheduled to available physical cores as demand changes. Physical CPU capacity that is not currently active can be used more efficiently across eligible VMs, while each DRVM can use up to its configured OCPU capacity.
- Dynamic memory management: DRVM uses memory overcommit while preserving the VM’s configured, guest-visible memory capacity. OCI optimizes the placement of DRVM instances across the fleet and continuously monitors host memory pressure. When needed, OCI rebalances eligible instances without rebooting them to help maintain performance.
Zero-downtime live conversion
Customers can update an eligible existing Standard VM and change its resource-management mode from Static to Dynamic. Eligible instances are converted live without a reboot. The instance retains its identity, configured resources, storage attachments, and network configuration.
Availability at launch
Dynamic Resource VMs are initially available for eligible VM.Standard.E6.Flex configurations in the following OCI regions: US East (Ashburn), US West (Phoenix), Germany Central (Frankfurt), UK South (London), US Midwest (Chicago), US West (San Jose), Saudi Arabia West (Riyadh), and Australia East (Sydney).
Customers can create eligible DRVM instances through the OCI Console, CLI, API, or Terraform. Capacity reservations can also be configured as DRVM for eligible shapes. Existing capacity reservations cannot be converted to DRVM. For current availability and configuration details, see DRVM documentation.
Getting Started
To create a DRVM in the OCI Console:
- Open Create compute instance in a supported region.
- Select an eligible VM.Standard.E6.Flex configuration.
- Configure the required OCPUs and memory.
- Under resource management, select Dynamic.
- Complete the remaining instance configuration and launch the VM.
For API and CLI deployments, set shapeConfig.resourceManagement to DYNAMIC. For Terraform deployments, set shape_config.resource_management = “DYNAMIC”.
To convert an eligible existing VM, edit the instance and change its resource-management mode from Static to Dynamic.
Review the DRVM documentation for supported configurations and detailed instructions.