Computing Platform – Data Centers with Carbon Innovative

Computing Platform

Google’s Technical Lead for its Carbon-Smart Computing Platform has made it clear that addressing the challenge of climate change requires a transformation in how the world produces and uses energy, especially in hyperscale data centers.

At Google, according to the company, it has been carbon neutral since 2007. However, 2019 marked the third consecutive year in which it matched its annual electricity consumption with 100% renewable energy purchases.

The company has also stated that it is working toward obtaining carbon-free energy 24 hours a day, seven days a week, in all locations where its data centers deliver products and services to billions of people worldwide.

This article belongs to the Cloud & IT.

Innovative Computing Platform

“To achieve 24×7 carbon-free power, our data centers need to work more closely with carbon-free power sources like solar and wind.”

Google has developed an intelligent carbon-aware computing platform as part of its sustainability efforts. This breakthrough was developed by a small team of engineers who identified opportunities to reduce the environmental impact of computing workloads.

The engineers designed and implemented this system for hyperscale data centers, which are facilities that support large-scale computing operations. The platform can shift the timing of flexible computing tasks to periods when low-carbon energy sources, such as wind and solar, are more abundant.

This process is designed to work without additional hardware and without affecting the performance of Google services such as Search, Maps, and YouTube, which people use around the clock.

Carbon-Free Energy

Google’s technical team has explained that shifting non-urgent computing tasks, such as creating new filter functions in Google Photos, processing YouTube videos, or adding new words to Google Translate, can help reduce the carbon footprint associated with electricity consumption. This approach supports the broader goal of moving toward carbon-free energy around the clock.

Computing Platform

Each day, Google’s Smart Carbon Platform compares two types of forecasts for the following day. One forecast, provided by its collaborator Tomorrow, estimates how the average hourly carbon intensity of the local electricity grid will change throughout the day.

A complementary internal forecast estimates the hourly energy resources a data center will need to complete its computing tasks during the same period.

By combining these forecasts, the system can identify opportunities to schedule flexible workloads when cleaner electricity is more readily available.

Optimizing Energy Use

Google uses these forecasts to optimize computing schedules hour by hour, aligning selected tasks with periods when low-carbon electricity is more abundant.

According to the company, early pilot results indicated that shifting computing jobs based on carbon intensity could increase the amount of low-carbon energy consumed by its data centers.

The pilot data also demonstrated how computing workloads could move away from a baseline schedule and toward times with lower carbon intensity, such as early morning or late evening, when wind power may be more abundant in certain locations.

This approach demonstrates how software-based scheduling can contribute to sustainability without necessarily requiring additional computing hardware.

Flexible Computing Tasks

The first version of this intelligent carbon-aware computing platform focuses on shifting flexible tasks within the same data center at different times of the day.

However, computing tasks can also potentially move between different data centers, allowing more work to be completed when and where electricity is cleaner.

Google’s longer-term approach involves changing both the timing and location of suitable computing workloads to maximize reductions in carbon emissions across its data center network.

The effectiveness of this approach depends on factors such as workload flexibility, data center capacity, network requirements, electricity availability, and service performance commitments.

Data in a Public Cloud

A public cloud is an infrastructure in which computing resources are shared among multiple users. However, the word “public” does not mean that the service is free or that user data is accessible to everyone. Cloud providers implement security mechanisms to protect information and manage access rights.

In a public cloud configuration, information is stored and processed in data centers operated by the cloud provider. These facilities support a wide range of services, from online storage and business applications to large-scale analytics. As organizations create and process growing amounts of information, understanding data capacity becomes increasingly important. For example, a yottabyte is a unit equal to (10^{24}) bytes in the decimal system, illustrating the enormous scale that digital information can reach.

The geographical location of data centers is also important for regulatory and privacy reasons. Legal requirements for handling information can differ depending on the country where data is stored or processed and the laws that apply to the provider.

For example, the US CLOUD Act can require covered service providers to disclose data within their possession, custody, or control in response to valid legal process, subject to applicable legal rules. Cross-border data access and transfers can therefore raise questions for organizations that must comply with European data protection requirements or other local regulations.

Cloud providers differ in how transparently they explain their data storage locations and processing arrangements. Online storage services such as Google Drive and Dropbox offer cloud-based storage, but customers should review the provider’s documentation and contractual terms to understand applicable data locations, access arrangements, and privacy safeguards.

Even when a provider operates data centers in Europe, customers should not automatically assume that every part of their information remains exclusively within a particular country. Data replication, backups, support operations, and other service arrangements may affect where information is processed.

Organizations should evaluate data residency options, contractual commitments, encryption, access controls, and compliance requirements before choosing a public cloud service.

Data in a Private Cloud

A private cloud is designed for use by a single organization, with computing resources dedicated to that organization rather than shared as a multi-tenant public cloud service. The organization may own and manage the infrastructure itself or use an external provider to operate it.

Private cloud computing can offer greater control over infrastructure, network access, and data management. Depending on how it is configured, this can help organizations implement specific security policies and address regulatory requirements.

However, a private cloud is not automatically more secure than a public cloud. Its security depends on the quality of its architecture, maintenance, access controls, monitoring, and operational procedures.

Private clouds can also provide greater control over capacity planning and system configuration. Organizations can design their environments around their own workloads and business requirements while limiting access to authorized users.

There are two common scenarios: the company owns the infrastructure, or it rents dedicated infrastructure from a service provider. The second arrangement is often called a hosted private cloud. Providers such as OVHcloud and Amazon offer different cloud services, although organizations should verify whether a particular service meets their requirements for a dedicated private environment.

More About Private Cloud

When an organization owns and operates its private cloud infrastructure, the location of its servers is generally easier to identify because the equipment is under its direct control. In a hosted private cloud, however, the customer should confirm the physical location of the infrastructure and the provider’s policies for backups, disaster recovery, and data transfers.

Some providers offer options to select a data center or geographic region. A dedicated infrastructure can give an organization more control over its environment, but the precise level of isolation and responsibility depends on the service agreement.

To use an analogy, renting a hosted private cloud is like renting a studio in a building. The owner manages the building itself, while the tenant has a dedicated space to arrange according to their needs. The tenant’s actual rights and responsibilities depend on the rental agreement.

Data localization and security remain important considerations whenever an organization relies on a service provider. Businesses should demand clear information about infrastructure, technology, security controls, data residency, and responsibilities for protecting confidential information.

Securing Cloud Computing Platforms

Whether an organization uses a public cloud or a private cloud, protecting data requires a comprehensive security strategy. Businesses should apply strong authentication, restrict access according to job responsibilities, maintain backups, monitor suspicious activity, and regularly review security configurations.

Secure data transmission is another important part of this strategy. Encryption and appropriate network protections help safeguard information as it moves between users, applications, devices, and cloud environments. Organizations should also consider how data is protected while stored and how access is managed across different systems.

In addition, companies with employees accessing cloud resources from smartphones, tablets, and laptops may benefit from mobile device management (MDM). MDM solutions can help administrators configure devices, enforce security policies, manage approved applications, and reduce the risk of unauthorized access from compromised or lost devices. MDM should complement, rather than replace, cloud security controls.

Ultimately, choosing the right computing platform requires balancing cost, scalability, performance, privacy, and environmental impact. By understanding the differences between public and private clouds and adopting appropriate security and sustainability practices, organizations can build more reliable and responsible digital infrastructure.