Qualcomm's Durga Malladi: HBC vs HBM, Dragonfly AI 250, and Winning on TCO

High Bandwidth Compute (HBC) serves as a critical technical differentiator for Qualcomm’s data center strategy, addressing the growing "memory wall" where compute capabilities outpace memory bandwidth. By utilizing near-memory computing, HBC performs processing directly adjacent to stacked DRAM, significantly reducing latency and power consumption. This architecture, debuting in the 2027 Dragonfly AI 250, enables massive performance gains in AI inference and token generation without increasing power envelopes. Durga Malladi, EVP and GM of Tech Planning at Qualcomm, emphasizes that this approach optimizes Total Cost of Ownership (TCO) by maximizing tokens per second per watt. Beyond data centers, this scalable technology aims to support agentic AI across edge devices, leveraging Qualcomm’s established supply chain relationships and manufacturing expertise to ensure production at scale.