Welcome to the next installment of the Bring Your Own Cloud blog series!
The first blog of this series introduced Ververica’s Bring Your Own Cloud Deployment option, offering a high-level overview of why BYOC was created and the benefits users can expect when utilizing this deployment option. The second blog explored “The Dilemma” businesses encounter when trying to find an ideal balance to ensure that a chosen deployment provides the appropriate level of flexibility, efficiency and security required. The third blog (split into Part One and Part Two) focused on security, dissecting the Zero Trust design choices Ververica made that future proof Ververica’s Unified Streaming Data Platform BYOC deployment.
This fourth piece tackles efficiency, another component of “The Dilemma,” by exploring how Ververica’s BYOC offering helps businesses to optimize Operational Expenditure (OPEX) and reuse Capital Expenditure (CAPEX) when building data streaming and processing projects.
Figure 1: The Dilemma: How to choose a deployment that best balances the needs of security, efficiency, and flexibility.
BYOC operates by clearly separating the vendor-managed control plane from the distributed data plane that resides in the customer’s cloud or region of choice. In this model, Ververica manages only the metadata, while your business retains complete control over your data (see Figure 2).
Figure 2: Ververica's BYOC control and data plane architecture
The beauty of a BYOC deployment lies in its cloud-native design, allowing it to work seamlessly with your existing infrastructure. This means that if you are already using cloud-native tools or resources, BYOC integrates smoothly, preserving prior investments and infrastructure.
In reality, businesses often have existing strategic partnerships with cloud providers (see Figure 3) including annual pricing plans and service commitments. They have likely already built a cloud-native infrastructure to run containerized applications orchestrated by Kubernetes, the de facto portability standard for running cloud-native apps. It’s also probable that specific compute instances, storage types, databases, and other resources are already in place and being utilized.
In many cases, these decisions are shaped by architectural constraints, especially during the lift-and-shift phase of a cloud-native journey, and based on required storage characteristics or the need for larger virtual machines to handle bigger workloads. For fully cloud-native architectures, the focus is often how to optimize OPEX, minimize footprint, increase density, and manage networking costs efficiently.
Figure 3: Choose from any Virtual Machine (VM) instance type and savings plan that aligns with your current business investments. (Sample of AWS options shown here.)
To allow for this flexibility, Ververica’s BYOC data plane software seamlessly integrates and runs in any type of compute, storage, and networking infrastructure underneath a Kubernetes and container runtime cluster.
In addition, you can create or decommission development and testing environments on-demand per team or per task, or share and continuously run them across teams. For on-demand environments, organizations can choose to decommission them during off-peak hours, and in the case of shared environments, workloads can be shut down when not in use to optimize costs. Ververica’s BYOC deployment supports both use cases, where rapid provisioning and de-provisioning of streaming workspaces is common for efficient development and testing.
Figure 4: BYOC allows utilization of existing investments, improving CAPEX and OPEX.
With Ververica’s Unified Streaming Data Platform BYOC deployment option, you can use:
You are probably familiar with Pay-as-you-go usage models that provide flexibility and cost efficiency for cloud offerings.
A key advantage of Ververica’s BYOC is its hourly prorated billing and unified payment system, which allows centralized cost tracking and real-time OPEX optimization across all BYOC workspaces. Businesses often overlook the significance of this level of cost control.
One of the biggest OPEX challenges in stream processing is the cost of both public and private cross-account network bandwidth. Ververica eliminates these costs by keeping stream processing and data movement within the customer’s own cloud and network domain, avoiding any cross-cloud account public or private link costs. This not only avoids expensive cross-account traffic fees, but also reduces operational complexity and maintenance costs associated with cross-account network peering and routing.
By running stream processing inside your network, Ververica ensures cost efficiency, performance optimization, and simplified operations. In short with Ververica, you can:
Kubernetes orchestration with container runtime has become the standard for cloud-native workload portability. Co-locating workloads within existing infrastructure enhances efficiency by reducing overhead and optimizing resource utilization. Ververica’s BYOC deployment follows this principle by ensuring its data plane microservices are lightweight, independent, and resilient. Additionally, you can co-locate multiple BYOC workspaces within the same compute and network infrastructure, enabling multi-tenant efficiency across different business units or teams within an organization. This approach allows Ververica to seamlessly integrate with existing workloads, maximizing Kubernetes cluster investments, and leveraging Kubernetes’ scheduling to achieve high compute density by running multiple workspaces (tenants) within the same compute capacity domain.
As shown in Figure 5, Ververica’s cloud-native data plane microservices can integrate with your existing compute, network, and storage infrastructures, as well as Kubernetes clusters to seamlessly co-locate with other workloads.
Figure 5: Integrate with K8s and Infrastructure-as-a-Service (IaaS)
In summary, Ververica’s BYOC offers:
In conclusion, Ververica’s BYOC deployment helps businesses to optimize OPEX and CAPEX when building data streaming and processing projects, providing a more efficient data streaming and processing deployment. This deployment optimizes efficiency at every level, offering advantages including:
Organizations know their cloud environments and infrastructure best, and Ververica’s Unified Streaming Data Platform BYOC deployment provides a highly flexible solution that allows businesses to pull the most appropriate cost-saving levers based on their unique operational needs.
Catch up on the entire BYOC Blog Series: