The New Era of Smartphone Storage

By: Christopher Bergey, Vice President of Embedded and Integrated Solutions, Western Digital

Storage performance and capacity within smartphones continued to increase significantly as smartphones have become the go-to devices for consumers globally. As such, mobile manufacturers around the world have nearly doubled device storage capacity on an annual basis as high-end smartphones now offer more capacity (256GB) than some laptops.

The need for increased smartphone performance was also a trend in 2016 as consumers captured and accessed significantly more of their photo and video memories enabled by these higher capacities and increased read/write speeds and transfer rates available from smartphone memory cards and mobile memory. The emergence of virtual reality (VR) and augmented reality (AR), 3D photography and 4K Ultra High Definition (UHD) video has also changed how smartphone content is consumed requiring processors and memory to function at much faster rates. With Internet traffic expected to grow three-fold over the next five years1, downloading a UHD movie could exhaust up to 80GB of storage capacity requiring faster networks and edge devices to store and access content locally.

As it relates to mobile trends for 2017, a faster 5G standard, even larger storage capacities for smartphones, and more prominent edge-to-cloud storage models are at the top of this list.

5G is Finally Closer to Reality
The mobile industry has been talking about 5G being the next telecommunications standard for quite some time. In addition to delivering faster Internet connection speeds and improved throughputs, 5G delivers higher user capacity than current 4G networks enabling more mobile broadband users per area and virtually unlimited data consumption. Users will be able to stream high-definition media many more hours per day with their mobile devices, and if Wi-Fi hotspots are out of reach, 5G will deliver improved device-to-device communication support, lower latency and less battery consumption than 4G platforms.

Though smartphones are the most connected devices that we use, as it relates to 5G, it is actually more about connecting to ‘other’ Internet devices with sensors. 5G will go well beyond smartphones and into computing devices that are embedded into everyday objects enabling them to capture, send and receive data. The Internet of Things (IoT) technology market is expected to grow from over $130 bn in revenue (USD) to more than $883 billion at a CAGR of 32.4% between 2016 and 2022, and covers hardware, platforms, software solutions and services that will create a host of derivative 5G products with enhanced storage capabilities.

1TB Capacity is Coming
Computing power in smartphones continues to improve as application processors take advantage of advancements in process technology and microarchitectures. Storage is following the same model. Today, 256GB of embedded storage is now available in smartphones (can actually extend to 512GB through a combination of embedded storage and microSDs). That’s a lot of storage capacity which a few years ago may have been deemed as excessive and unnecessary. However, consumers today never expect to delete content from applications, images, videos, documents or games. If virtual reality is added to this mix, the size of files required to run in smartphones will be even more astounding. Therefore, it is necessary for storage to scale just like processing power.

What we are seeing is a miniaturization of storage. For example, Western Digital’s 256GB microSD cards are as small as a thumbnail, extremely power efficient and essential for mobile devices with limited real estate like cameras, wearables, and of course, smartphones. The industry is not far away from having 1TB storage capacities in smartphones and know that consumers will not only want, but demand, terabyte capacities in their mobile devices.

Edge-to-Cloud Storage
In many mobile devices from self-driving cars to drones to smartphones, machine intelligence is typically built around computer vision and a variety of sensors because the priority is typically focused on superlative imaging, thus placing higher emphasis on camera differentiation. Many consumers assume that all of this data associated with machine intelligence will simply go to the cloud, which is incorrect as the new IoT model is far more complex.

On one hand, the cloud is elastic, ubiquitous and has analytical tools that may help discover hidden secrets in captured data. On the other hand, real-time analytics cannot be performed on cloud-based data due to bandwidth challenges and the amount of data being captured. By the time data gets into the cloud, it is historical, eliminating timely responses associated with real-time information.

Since there are many situations that require instantaneous responses, there is an urgency for real-time analytics to run near the edge of the cloud (or fog as it is deemed) in the form of gateways or aggregated systems. When this occurs, mobile devices will now be equipped to analyze and respond in real-time to a variety of alerts. As better data intelligence and pattern recognition tools are becoming increasingly necessary across all areas of IoT, a deeper understanding of the inherent value in captured data, whether in the cloud or at the edge, will be realized.

Therefore, much of the data and computing will happen at the edge, rather than in the cloud. There will be some sharing of that data that will move into the cloud in the form of machine learning or artificial intelligence, however, in this scenario, edge storage and edge compute models will complement the cloud by capturing data and making local decisions based on algorithms within the mobile device so that the cloud can actually be used for learning. As a result, the decision algorithms are passed down to edge devices for local decisions enabling mobile users to learn what they are doing well and not doing well requiring more storage at the edge and fog.

From 5G becoming reality to mega-smartphone capacities to edge-to-cloud storage models, 2017 is shaping up to be an exciting year for data storage in the mobile industry. In fact, given the emphasis that OEMs are now placing on differentiating the CMOS sensor in mobile devices to produce superlative imaging, storage is now at the top of mind for many consumers who have become more conscientious that their mobile devices have adequate storage for storing and sharing. It’s not just capturing 2D pictures any longer, but it’s now 3D pictures, virtual reality, augmented reality, video, and richer and richer content in general, which means that files have become larger, requiring a lot more storage, but more importantly, high-performance storage.