Getty Images/iStockphoto


3 factors that influence multi-cloud cost optimization

Without careful oversight, multi-cloud deployments can be expensive. These data management and security practices can help IT teams control and reduce monthly costs.

A multi-cloud strategy offers access to a breadth of cloud services and a reduced risk of vendor lock-in. Along with their many benefits, multi-cloud deployments can also introduce extra costs that are not present in single-cloud environments.

To keep costs in check, IT teams should understand application requirements and track data usage and networking trends. They must also consider broader obligations, such as security and compliance.

Multi-cloud cost considerations

An example helps illustrate the specific multi-cloud costs an enterprise can incur.

Let's say an organization runs an application in AWS that generates transaction data. Then, the company moves that data to Google Cloud to train machine learning models. In this multi-cloud scenario, numerous factors could affect -- and ultimately increase -- costs, including:

  • Whether transaction data is stored on both AWS and Google Cloud.
  • Data lifecycle management practices across clouds.
  • Current network configurations, plus bandwidth and latency requirements.
  • Access controls and data-loss prevention measures to protect data in each cloud, as well as data in transit.

Factors of multi-cloud cost optimization

Take a closer look at these three factors and how they play a role in multi-cloud cost optimization.

1. Data lifecycles

When organizations have customer-facing services that generate data, or accumulate data for analytics, they must pay close attention to multi-cloud storage costs.

Apply different data lifecycle management practices to different data types. For example, store used transaction data and snapshots close to the services that use them. This reduces latency and boosts performance. To mitigate the risk of service failure, teams can also store copies of data in different cloud regions or availability zones.

Store older transaction data that is not frequently accessed in lower-cost storage services, such as archives. These services do not have to live on the same cloud platform that hosts the application that generates the data.

Implement proper data lifecycle management
Follow the lifecycle of data in this flowchart.

Use policy management and data migration services to automatically move data from high-cost, low-latency storage systems to lower-cost storage. These lower-cost systems can provide long-term backup and act as a repository for data analytics and machine learning. Consider whether copies of data in one long-term, low-cost cloud storage service are sufficient or whether copies will be necessary in multiple clouds.

2. Data synchronization

Data synchronization is another factor that plays into multi-cloud cost optimization.

The best practices mentioned above are useful for long-term storage operations. When IT teams need to synchronize changes to data across clouds -- and do so with minimal latency -- they must take alternative approaches.

For example, a cloud-hosted application that processes financial transactions may need to send data to a fraud-detection service that runs in another cloud. This would require a reliable service to receive the data and buffer it until the fraud-detection service can process it. IT teams could use a message queue service, such as Amazon Simple Notification Service or Google Cloud Pub/Sub, to ingest and store the data until it is processed.

Consider the reliability requirements of multi-cloud services. Can the system or workload function with some missing or delayed data? If not, high reliability and low latency are required, and IT teams might need to deploy additional networking services to meet these demands.

If lower durability and availability are tolerable, consider networking services such as Google Cloud Platform Pub/Sub Lite. These cost less than the full-featured version of the service mentioned above, and they aid in multi-cloud cost optimization.

3. Identity and access control management

When it comes to multi-cloud cost optimization, data protection and compliance aren't always the first items to come to mind for IT teams. But a lack of focus on security and compliance requirements can result in big expenses.

Benefits of IAM tools in cloud

When using data across multiple clouds, security becomes complex. The time and effort required to secure data in one cloud can double when admins must take similar measures in a second cloud. To reduce redundant work, use a single sign-on service that allows for federated identities. Assign attributes to identities, such as users and service accounts. This determines the privileges available to them. Use identity and access management tools to help with this task. Unify identity management and authorization policies to reduce redundant security management operations across clouds.

In addition, when copying data from one cloud to another, use data-loss prevention services to scan and redact protected data that is not needed by services running in the second cloud.

Dig Deeper on Cloud infrastructure design and management

Data Center