Data Centers Need Green Storage to Meet Carbon Peak & Neutrality Goals  

Green data storage is essential for data centers to reach net-zero carbon emissions. Organizations around the world are working to meet carbon peak and neutrality goals, and this starts with data centers. More than 30% of a data center's energy consumption goes to storage. Therefore, to build sustainable data centers, we need to focus on reducing the energy consumption of IT equipment, particularly, storage devices, in addition to lowering power usage effectiveness (PUE).

Trends

Building a green data center requires energy-efficient data storage as well as lower PUE

Reduced PUE is just one step along the way to green data centers. Lowering the power consumption of IT equipment is arguably even more important. Storage devices are expected to be the main electricity-draining IT components. For example, the annual power consumption of 1 petabyte of data storage in a data center is 300,000 kWh, which is equivalent to emitting 235.5 tons of carbon. Without an effective green strategy, the carbon emissions attributed solely to storage in 2030 will easily exceed the total global carbon emissions recorded in 2019.

Figure 1: Lifecycle carbon footprint

Innovations in energy-saving storage technologies are a catalyst for the low-carbon development of data centers

In response to the mounting pressure to reduce storage energy consumption, storage vendors are proactively innovating and developing technologies to help data centers go green and contribute to sustainable development.

Data reduction technology squeezes as much data as possible into physical storage, without distorting it, to maximize capacity. This lowers operation costs for enterprises and reduces the amount of energy data storage devices consume.

High-density design improves storage capacity density.

Storage media accounts for 83% of a device's total power consumption. Compared with HDDs, SSDs consume 70% less power and save 50% space with the same capacity. A storage product equipped with large-capacity SSDs and high-density disk enclosures can store the same amount of data with less energy and space, meaning lower power consumption per unit capacity.

Near-storage computing reduces the energy consumed by data movement.

Data movement increases energy consumption. Research shows that data movement consumes almost twice as much energy as computing on large-scale AI computing clusters. However, the energy it consumes on local compute nodes is only 5% of that of computing. Processing data close to where it resides reduces data movement. For example, processing tasks like vector retrieval on storage helps halve overheads in protocol and data conversion between storage and computing, and more than halve the energy required for data processing.

Accelerated dedicated processors reduce CPU overheads.

Compared with storage devices with a single CPU computing architecture, devices equipped with dedicated processors perform better and use less power. Dedicated processors that take over data reduction and protocol processing tasks from general-purpose CPUs boost efficiency and reduce latency for data processing, and eliminate dependencies on CPU computing power, thereby making the storage solution more energy efficient.

Load-based intelligent frequency modulation and core adjustment slashes power consumption.

Predictions and interventions based on AI models and software solutions make storage operations in data centers more energy efficient. Big data analytics of service I/O models, technologies like intelligent load prediction, dynamic resource scheduling management, and the timely shutdown and adjustment of resource usage frequency enable storage devices to have the lowest energy consumption during operation while also meeting SLA requirements for service workloads without affecting upper-layer applications.

Energy saving through storage lifecycles.

Energy savings need to be made throughout the entire lifecycle of storage devices, including raw material selection, manufacturing, transportation, product use, and disposal. In the storage manufacturing phase, manufacturing plants widely use photovoltaic power, zero wave soldering, paperless labeling, and renewable materials such as aluminum and tin. Packaging made from FSC-certified paper and printed using soybean ink weighs less and so further eases the transportation burden. In the storage products' use phase, intelligent O&M based on AIOps enables on-demand use of storage resources. In addition, a proper recycling system is established to ensure that at the end of the product lifecycle, e-waste is handled in an environmentally friendly manner that optimizes recycling and minimizes environmental impact.

The current spotlight on carbon footprint is motivating the optimization of storage product design and the development of energy-saving technologies. With its industry-leading innovative green solutions, Huawei OceanStor All-Flash Storage has become the first to be awarded DEKRA Storage Product Carbon Footprint and DEKRA Seal certificates. DEKRA is one of the world's leading expert organizations in the testing, inspection, and certification sector.

Suggestions

Shift focus from the current power consumption of a single device to carbon emissions throughout the device's entire lifecycle

An increasing number of global enterprise customers require equipment suppliers to provide carbon footprint reporting or carbon emissions reporting for the entire lifecycles of their products. As carbon disclosure continues to draw widespread attention, carbon footprint reporting is becoming a mainstream requirement as we work towards a green future in international trade. Currently, carbon footprint evaluation standards and its accounting methods vary across the industry. The best course of action would be for enterprises to use certification from reputable third-party testing organizations.

Comply with unified energy efficiency evaluation standards

Standard organizations, enterprise customers, and storage vendors use different methods and indicators to assess how energy efficient storage solutions are because the field is still in its infancy. This means enterprises and vendors repeat tests based on various standards and evaluation methods. The establishment of a unified set of energy efficiency standards can reduce both resource waste and pressure on vendors, who have to get their products repeatedly tested. It could also help enterprise customers make more informed choices when it comes to green storage solutions, and facilitate the healthy development of energy-saving and low-carbon technologies in the storage industry.

Promote innovation by storage vendors to lower power consumption

Enterprises are encouraged to proactively deploy storage products powered by energy-saving technologies and push storage vendors to innovate, for example, in dedicated processors, data reduction, high-density all-flash design, and intelligent frequency modulation for higher capacity density and energy efficiency. Next-generation storage products can be equipped with near-storage computing to reduce the amount of energy used for data movement.

Learn more about Huawei Storage and subscribe to this blog to get notifications of all the latest posts.

Disclaimer: Any views and/or opinions expressed in this post by individual authors or contributors are their personal views and/or opinions and do not necessarily reflect the views and/or opinions of Huawei Technologies.