Home/ ECE/ Chiplet-Based 3D IC Stacking
ECE · Seminar 09 · Building big chips from small dies

Chiplet-Based 3D IC Stacking

Chiplet integration assembles multiple small dies into one package, with 3D stacking and advanced interconnects overcoming the limits and yield loss of ever-larger monolithic chips.

chiplet3D ICadvanced packagingUCIeTSVheterogeneous

Moore's-law scaling of single monolithic chips is slowing and increasingly expensive: huge dies have poor yield and mixing process nodes is wasteful. The chiplet approach disaggregates a system into smaller dies — each built on the optimal node — then re-integrates them in one package using advanced packaging and 3D stacking.

Working principle

Chiplets are connected with very short, dense links. 2.5D integration places dies side-by-side on a silicon interposer or bridge; 3D integration stacks dies vertically and connects them with through-silicon vias (TSVs) and hybrid bonding, drastically cutting interconnect length and energy. Open die-to-die standards such as UCIe let chiplets from different vendors interoperate.

Stacked logic / cache die3D-bonded via TSV / hybrid bondL4Compute chiplet (advanced node)CPU/GPU/AI tilesL3I/O & analog chiplet (mature node)PHYs, SerDes — cheaper processL2Silicon interposer / substrateDense routing, UCIe die-to-die linksL1Heterogeneous 2.5D/3D chiplet package
Figure 1. Each function uses its best-fit process; an interposer and TSVs knit the dies into one high-bandwidth system.
Table 1. Monolithic SoC vs. chiplet design
PropertyMonolithic SoCChiplets
YieldFalls with die sizeHigh (small dies)
Process mixingSingle nodeBest node per function
ReuseLowHigh (mix & match)
InterconnectOn-dieDie-to-die (UCIe, bridge)
ChallengeCost at large sizePackaging, thermals
Key challengeHeat removal and the die-to-die interface dominate chiplet design; 3D stacks concentrate power density, making thermal-aware floorplanning essential.

Applications

  • High-performance CPUs/GPUs and AI accelerators (HBM-on-package)
  • Cost-optimised SoCs mixing advanced and mature nodes
  • Heterogeneous integration of compute, memory and photonics

References & further reading

  1. Naffziger et al., “Pioneering Chiplet Technology … AMD EPYC and Ryzen,” ISCA 2021.
  2. UCIe Consortium, “Universal Chiplet Interconnect Express Specification,” 2022–2024.
  3. Lau, “Recent Advances and Trends in Heterogeneous Integration Packaging,” IEEE Trans. CPMT, 2022.