Explaining the QSFP-DD Data Center Interconnect Standard

Estimated reading time: 1 minute

The explosion in data center activity shows no signs of abating. Cisco’s latest Visual Networking Index (VNI) study projects annual global IP traffic will reach 4.8 zettabytes per year by 2022 after breaking one zettabyte just three years ago. Worldwide traffic will triple over the next five years, reaching 50 GB per capita. The VNI study anticipates 28.5 billion networked devices by 2022, up from 18 billion in 2017, with 82% of all data flow due to video viewing.

According to Cisco’s report, most IP traffic either originates or terminates in data centers. Yet as massive as the flows between centers are becoming, the data managed within those same facilities is going even higher. All of this activity means that “hyperscale” is the word of the future. Roughly one-quarter of all servers installed in 2016 went to hyperscale facilities, but that number will grow to almost one-half by next year. Other advances that are heavily impacting the modern data center is the rise in 400-Gbps technology, the expansion of cloud computing, and how virtual data centers, machine learning, and AI are all driving change.

To meet the pressing need for high signal integrity, lower latency, lower insertion loss, and higher density data transmission, transceivers and modules capable of supporting 400 Gbps are quickly being developed. At the forefront is the Quad Small Form Factor Pluggable-Double Density (QSFP-DD) system— the most advanced interconnect standard yet. Defining form factors and performance specifications for this new standard is the QSFPDD Multi-Source Agreement (MSA) comprised of leading data center equipment and component suppliers.

To read this entire article, which appeared in the September 2019 issue of SMT007 Magazine, click here.