SK hynix Develops High Bandwidth Storage (HBS) to Eliminate Mobile AI Bottlenecks Using VFO Technology

SK hynix introduces High Bandwidth Storage (HBS), a technology using vertical wire fan-out (VFO) packaging to stack DRAM and NAND, solving mobile AI
SK hynix Develops High Bandwidth Storage (HBS) to Eliminate Mobile AI Bottlenecks Using VFO Technology

SK hynix Combats Mobile AI Bottleneck with High Bandwidth Storage (HBS)

The South Korean major memory producer, SK hynix, is developing a new technology called High Bandwidth Storage (HBS) to address the performance bottlenecks of mobile devices. This technology will allow for more powerful on-board AI power for future smartphones and tablets by packaging DRAM and NAND stacked with vertical wire fan-out (VFO) technology.

The Role of Vertical Wire Fan-Out (VFO) Packaging

The success of HBS resides in the vertical wire fan-out (VFO) packaging technology developed by SK hynix and introduced in 2023. VFO differs from conventional wire bonding in that it is designed to connect chips through straight lines of stacked DRAM and NAND. The difference in many ways is important:

  • Less wiring distance.
  • Insignificant signal transmission loss or delay than in most cases.
  • Allows for a larger number of I/O connections.

The improvement will go toward rising data processing capability more in need of some developments at the same time when on-chip generative AI becomes a standard on a smartphone.

SK hynix Develops High Bandwidth Storage (HBS) to Eliminate Mobile AI Bottlenecks Using VFO Technology

HBS Applications and Cost Glory

To implement High Bandwidth Storage, it will be integrated onto the main logic board of the smartphone by being fitted combined with the System-on-Chip (SoC). Although specifically onboarding SoCs with HBS is not signed yet, it might be beneficial for rumored Snapdragon 8 Elite Gen 6 Pro chip, which is rumored to come with LPDDR6 and UFS 5.0 support.

A major advantage High Bandwidth Storage offers over HBM is in its manufacturing. HBS achieves this at a cost-efficient process that bypasses TSV, which basically charges for piercing through the works of a chip at an enormous cost. It results in:

  • Low-cost fab.
  • More die per wafer as yield gets better.

With such a paradigm effect in cost structures, adding them seems unbeatable for phone and tablet makers. If Apple is surely looking at HBM with TSV to operate heavy AI models on advanced iPhones in the future, a more cost-effective option like HBS would get high traction and put into use very quickly.

About the author

mgtid
Owner of Technetbook | 10+ Years of Expertise in Technology | Seasoned Writer, Designer, and Programmer | Specialist in In-Depth Tech Reviews and Industry Insights | Passionate about Driving Innovation and Educating the Tech Community Technetbook

Join the conversation