YZYang Zhang

Yang Zhang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yang Zhang has a total of 9 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Power Delivery Network Modeling and Benchmarking for Emerging Heterogeneous Integration Technologies

    by Yang Zhang, Obaidul Hossen, Muhannad S. Bakir

    In this article, a power delivery network (PDN) modeling framework for heterogeneous integration platforms is presented, which includes both IR-drop and transient analysis. The model is validated using IBM power grid benchmarks, and maximum relative errors are less than 7.3%. To evaluate interposer and bridge-chip-based integration platforms, we assume a field-programmable gate array (FPGA)-CPU 2.5-D integration platform, of which the FPGA consumes 45 W and the CPU consumes 75 W. The simulation results show that an interposer with dense power/ground grid and microbumps can suppress power supply noise (PSN) by a small margin with the requirement of high-density through-silicon vias (TSVs). For bridge-chip-based integration, under the assumption that the active dice above the bridge-chips are not connected to the package power/ground planes, some PDN challenges are evaluated. Using multiple bridge-chips and smaller overlap areas between the bridge-chips and the active dice, the worst-case PSN in bridge-chip-based integration is minimized. Next, we perform parametric studies for these 2.5-D integration platforms as a function of C4 bump pitch and metal layers. Lastly, we propose to use TSVs in the bridgechip to reduce PSN. With nine bundled TSVs of 8 μm diameter, the IR-drop of bridge-chip-based integration technologies is only 1% higher than the standalone case.


  • Benchmarking Digital Die-to-Die Channels in 2.5-D and 3-D Heterogeneous Integration Platforms

    by Yang Zhang, Xuchen Zhang, Muhannad S. Bakir

    In this paper, compact circuit models and HSPICE simulations are used to benchmark die-to-die communication channels in 2.5-D and 3-D heterogeneous integration platforms. The delay, energy-per-bit, and bandwidth-density of the considered integration platforms are simulated and benchmarked. Compared to other 2.5-D integrated systems with a 1-mm interconnect length, heterogeneous interconnect stitching technology (HIST)-based 2.5-D integration shows a maximum latency and energy reduction of 6.2% and 15.1%, respectively. 3-D ICs show further performance enhancement compared to 2.5-D integration; the link latency and energy are approximately 19.4% and 48.0% smaller than those of HIST (1-mm wire) for through-silicon via (TSV)-based 3-D integration (75-μm TSV height). Next, the impacts of the physical I/O interconnect dimensions and device process technology scaling are evaluated and we observe that advanced process technologies must be integrated with smaller physical I/O dimensions and shorter wire lengths to attain full advantages of scaling. Finally, we consider the thermal impact of dense heterogeneous integration and investigate the thermal and electrical signaling tradeoffs in 2.5-D and 3-D integration.


  • Open Access

    Power Delivery Network Benchmarking for Interposer and Bridge-Chip-Based 2.5-D Integration

    by Yang Zhang, Obaidul Hossen, Muhannad S. Bakir

    In this paper, a power delivery network (PDN) modeling framework for emerging heterogeneous 2.5-D integration platforms is presented. The framework is validated using IBM power grid benchmarks and maximum relative errors of less than 7.29% and 0.67% for IR-drop and transient noise are shown, respectively. Next, the framework is used to evaluate interposer and bridge-chip based 2.5-D integration platforms. The simulation results show that an interposer with dense power/ground grids and microbumps can suppress power supply noise (PSN) by a small margin. In bridge-chip based 2.5-D integration, under the assumption that the bridge-chips underneath the active dice block direct access to package power/ground planes, some PDN considerations are highlighted and evaluated. Using multiple bridge chips and smaller overlap areas between the bridge-chips and the active dice, the worst-case PSN in bridge-chip based 2.5-D integration is minimally impacted.


  • Monolithic Integration of a Micropin-Fin Heat Sink in a 28-nm FPGA

    by Thomas E. Sarvey, Yang Zhang, Colman Cheung, Ravi Gutala, Arifur Rahman, Aravind Dasu, Muhannad S. Bakir

    Microfluidic cooling has been demonstrated as an effective means of cooling microelectronic circuits with a very low convective thermal resistance and potential for integration in close proximity to the area of heat generation. However, microfluidic cooling experiments to date have been limited to silicon with resistive heaters representing the heat generating circuitry. In this paper, a micropin-fin heat sink is etched into the back side of an Altera Stratix V field-programmable gate array (FPGA), built in a 28-nm CMOS process. Thermal and electrical measurements are made running a benchmark pulse compression algorithm on the FPGA. Deionized water is used as a coolant with flow rates ranging from 0.15 to 3.0 mL/s and inlet temperature ranging from 21 °C to 50 °C. An average junction-to-inlet thermal resistance of 0.07 °C/W is achieved.


  • Open Access

    Thermal Evaluation of 2.5-D Integration Using Bridge-Chip Technology: Challenges and Opportunities

    by Yang Zhang, Thomas E. Sarvey, Muhannad S. Bakir

    In this paper, 2.5-D integrated circuits (ICs) using bridge-chip technology are thermally evaluated to investigate thermal challenges and opportunities for such multi-die packages. To this end, the objectives of this paper are twofold. First, thermal benchmarking of a number of 2.5-D integration approaches is performed and compared to 3-D ICs for completeness. Thermal modeling shows that the evaluated 2.5-D integration approaches exhibit similar thermal characteristics, but show significant improvements compared to 3-D IC solutions with the same power consumption. Second, this paper explores bridge-chip-based 2.5-D integrated systems as a function of bridge-chip thickness, thermal interface material properties, microbump properties, die thickness, die thickness mismatch, and die-to-die spacing along with transient analysis to investigate time-domain thermal coupling. Results suggest that a die thickness mismatch of 100 μm can increase the maximum temperature by 25.8% Therefore, the die thickness mismatch should be kept as small as possible, and the hottest die should be the thickest. Moreover, reducing die-to-die space from 1 to 0 mm increases thermal coupling, and the low-power dice, such as DRAM, can have a temperature increase of 6.1%.


  • Open Access

    Integrated Thermal and Power Delivery Network Co-Simulation Framework for Single-Die and Multi-Die Assemblies

    by Yang Zhang, Muhannad S. Bakir

    This paper presents a thermal and power delivery network (PDN) co-simulation framework for single-die and emerging multi-die configurations that incorporates the interactions between temperature, supply voltage, and power dissipation. The temperature dependencies of wire resistivity and leakage power are considered, and the supply voltage dependencies of power dissipation are modeled. Starting with a reference power dissipation, the framework is capable of evaluating the temperature distribution and PDN noise simultaneously and eventually updating the power dissipation based on the thermal and supply voltage distributions. The simulation results of an example two-tier 3-D stack show that prior models considering only part of the interactions between temperature, supply voltage, and power dissipation have a maximum error of 7.66%, 9.79%, and 4.64% in IR-drop, transient power supply noise, and temperature, respectively.