Storj, the leader in enterprise-grade, globally distributed cloud object storage, today released the findings from its sustainability whitepaper on the environmental impact of cloud storage. The report, produced by the Storj team and reviewed by academic researchers, identifies the carbon costs of traditional cloud storage in centralized data centers and presents a model to estimate carbon savings from using Storj compared to hyperscaler and centralized clouds.
The demand for cloud storage is growing exponentially, as is the insatiable need for reliable, secure, and performant cloud storage. The rise of AI, which requires large datasets for machine learning, is exacerbating the problem. Centralized providers like Amazon AWS, Microsoft Azure, and Google Cloud rely on expanding data centers, which account for 3% of all electricity generated on Earth. According to Greenly, data storage centers are responsible for 2% of global carbon emissions when you include all devices that make use of data storage.
“When assessing the environmental impact of centralized providers, it’s essential to consider not only the electricity generated in powering and cooling data storage devices but the environmental impact from manufacturing and transporting those devices as well,” said JT Olio, Chief Architect at Storj. “Storj is able to eliminate much of those costs by storing data on the under-utilized capacity of drives that are already manufactured, powered, and cooled for other purposes. Through our analysis, we estimate that per terabyte of data, Storj’s distributed model reduced the carbon footprint of storing a TB of data by 66-83% over a three-year period compared to other providers.”
The report outlines significant carbon savings are additionally achieved by:
-
Extending the effective life of already manufactured drives.
-
Minimizing the number of copies of data that need to be stored to achieve equivalent data durability, global distribution, and performance levels.
-
Enabling a carbon-efficient distribution of storage between hard disk drives and solid state drives.
Storj leverages the underutilized storage capacity from over 24,000 nodes independently operated worldwide. 80% of disk drives worldwide are only 20% full; by leveraging this existing capacity, Storj eliminates the need to build new drives and data centers, while alleviating the need for incremental electricity and cooling. The distributed architecture is also designed to protect against outages, ransomware, and data compromise while boosting performance and reliability.
“We understand that customers have traditionally had to prioritize performance and cost first, sustainability second,” said Ben Golub, CEO of Storj. “Storj is proud to lead the trend where a sustainable solution doesn’t compromise performance and in fact can be 90% less expensive than leading cloud object storage providers. What we’ve done at Storj is leverage a distributed model to create a truly viable alternative to hyperscalers that is not only more cost-effective and sustainable but outperforms centralized platforms in performance, security, and reliability.”
About Storj
Storj is a leader in decentralized cloud object storage. Built for enterprise architects, IT ops professionals, and developers, Storj delivers blazingly fast, CDN-like performance at cold storage prices, enterprise-grade durability, and better security with no vendor lock-in and no single points of failure. Storj meets the leading-edge privacy and sustainability demands for traditional use cases, Web3, and dApps. Easily integrated into any existing stack with S3 compatibility, Storj is architected as a trustless globally distributed network that utilizes existing excess storage capacity making performance, privacy, and resiliency available to any size organization, at 1/5 to 1/20 the price of hyperscalers. Storj stores multiple petabytes and has petabyte-scale enterprise and Web3 customers and partners like Pocket Network, Atempo, iXsystems, Gabb Wireless, and the University of Edinburgh. STORJ is an ERC-20 token used across the Storj network.