Page 49 - ChipScale_Jul-Aug_2020-Digital
P. 49

Figure 4: Key drivers of SiP vs. standalone package. SOURCES: [1,2]
                                                                              References
                                                                                1.  “System-in-package technology and
                                                                                  market trends 2020” report, Yole
                                                                                  Développement, 2020.
                                                                                2.  “Advanced packaging technology in
                                                                                  the Apple Watch Series 4’s system-
                                                                                  in-package” report, System Plus
                                                                                  Consulting, 2019.
                                                                                3.  “Qualcomm QET5100M envelope
                                                                                  tracker module with SEMCO’s
                                                                                  e m b e d d e d d i e p a c k a g i ng
                                                                                  technology” report, System Plus
                                                                                  Consulting, 2019.
                                                                                4.  “NVDIA Tesla P100 GPU with
                                                                                  H BM 2” re por t, System Plu s
                                                                                  Consulting, 2017.
        Figure 5: SiP commercialized products. SOURCES: [3-5]                   5.  Advanced system-in-package
          For high-end SiP, the Nvidia Tesla   CPUs. Deep learning training (comparing   technology in Apple’s Airpods Pro”
        V100 GPU is meant for data centers and   8X V100 & 8X P100 modules) can take   report, System Plus Consulting, 2020.
        is used to accelerate artificial intelligence   5x less time than with the Tesla P100
        (AI), HPC and graphics. Tesla V100   GPU, and its deep learning inference   Acknowledgment
        c ome s i n 16 a nd 32GB me mor y   (comparing 8X V100 & 8X P100 modules   The authors wish to thank Stéphane
        configurations and 900GB/s total   & Xeon CPU) has 3X higher throughput   Elisabeth, PhD, Technology & Cost
        bandwidth, with power consumption of   than Tesla P100 and is 7X more powerful   Analyst at System Plus Consulting, for
        250-300W. In a single GPU, Tesla V100   than a Xeon CPU.              his contributions to this paper.
        offers performance equivalent to 100


                       Biographies
                         Favier Shoo is a Technology and Market Analyst in the Semiconductor & Software division at Yole
                       Développement, part of the Yole Group of Companies. Based in Singapore, he is engaged in the development
                       of technology and market reports, as well as the production of custom consulting reports. He holds a Bachelor’s
                       in Materials Engineering (Hons) and a Minor in Entrepreneurship from Nanyang Technological U. (NTU)
                       (Singapore). He was also the co-founder of a startup company. Email: favier.shoo@yole.fr
                         Santosh Kumar is currently working as Principal Analyst and Director Packaging, Assembly & Substrates for
                       Yole Développement’s activities in Korea. He is involved in the market, technology and strategic analyses of the
          microelectronic assembly and packaging technologies. He is the author of several reports on fan-out/fan-in WLP, flip-chip, and
          3D/2.5D packaging and received Bachelor’s and Master’s degrees in Engineering from the Indian Institute of Technology (IIT),
          Roorkee and U. of Seoul, respectively.


                                                                                                             47
                                                             Chip Scale Review   July  •  August  •  2020   [ChipScaleReview.com]  47
   44   45   46   47   48   49   50   51   52   53   54