Ankit Kaul
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Ankit Kaul has a total of 30 co-authors. Here are the top co-authors by number of shared publications:
Muhannad S. Bakir
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
6 shared publications- Shimeng Yu
Georgia Institute of Technology
6 shared publications Madison Manley
School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
3 shared publicationsMuhannad Bakir
Department of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, USA
3 shared publicationsAravind Dasu
Intel Corporation, Santa Clara, CA, USA
2 shared publications- James Read
Georgia Institute of Technology
2 shared publications
Latest Publications
- Open Access
- 27 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
- Vol. 11, 10-18
Co-Optimization of Power Delivery Network Design for 3-D Heterogeneous Integration of RRAM-Based Compute In-Memory Accelerators
3D heterogeneous integration (3D HI) offers promising solutions for incorporating substantial embedded memory into cutting-edge analog compute-in-memory (CIM) AI accelerators, addressing the need for on-chip acceleration of large AI models. However, this approach faces challenges with power supply noise (PSN) margins due to VDD scaling and increased power delivery network (PDN) impedance. This study demonstrates the necessity and benefits of 3D-HI for large-scale CIM accelerators, where 2D implementations would exceed manufacturing reticle limits. Our 3D designs achieve 39% higher energy efficiency, 8× higher operation density, and improved throughput through shorter vertical interconnects. We quantify steady-state IR-drop impacts in 3D-HI CIM architectures using a framework that combines PDN modeling, 3D-HI power, performance, area estimation, and behavioral modeling. We demonstrate that a drop in supply voltage to CIM arrays increases sensitivity to process, voltage, and temperature (PVT) noise. Using our framework, we model IR-drop and simulate its impact on the accuracy of ResNet-50 and ResNet-152 when classifying images from the ImageNet 1k dataset in the presence of injected PVT noise. We analyze the impact of through-silicon via (TSV) design and placement to optimize the IR-drop and classification accuracy. For ResNet architectures in 3D integration, we demonstrate that peripheral TSV placement provides an optimal balance between interconnect complexity and performance, achieving IR-drop below 10% of VDD while maintaining high classification accuracy.
- 27 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
- Vol. 11, 10-18
Open AccessCo-Optimization of Power Delivery Network Design for 3-D Heterogeneous Integration of RRAM-Based Compute In-Memory Accelerators
3D heterogeneous integration (3D HI) offers promising solutions for incorporating substantial embedded memory into cutting-edge analog compute-in-memory (CIM) AI accelerators, addressing the need for on-chip acceleration of large AI models. However, this approach faces challenges with power supply noise (PSN) margins due to VDD scaling and increased power delivery network (PDN) impedance. This study demonstrates the necessity and benefits of 3D-HI for large-scale CIM accelerators, where 2D implementations would exceed manufacturing reticle limits. Our 3D designs achieve 39% higher energy efficiency, 8× higher operation density, and improved throughput through shorter vertical interconnects. We quantify steady-state IR-drop impacts in 3D-HI CIM architectures using a framework that combines PDN modeling, 3D-HI power, performance, area estimation, and behavioral modeling. We demonstrate that a drop in supply voltage to CIM arrays increases sensitivity to process, voltage, and temperature (PVT) noise. Using our framework, we model IR-drop and simulate its impact on the accuracy of ResNet-50 and ResNet-152 when classifying images from the ImageNet 1k dataset in the presence of injected PVT noise. We analyze the impact of through-silicon via (TSV) design and placement to optimize the IR-drop and classification accuracy. For ResNet architectures in 3D integration, we demonstrate that peripheral TSV placement provides an optimal balance between interconnect complexity and performance, achieving IR-drop below 10% of VDD while maintaining high classification accuracy.
- Top Cited PapersOpen Access
- 23 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
- Vol. 10, 89-97
Heterogeneous Integration Technologies for Artificial Intelligence Applications
The rapid advancement of artificial intelligence (AI) has been enabled by semiconductor-based electronics. However, the conventional methods of transistor scaling are not enough to meet the exponential demand for computing power driven by AI. This has led to a technological shift toward system-level scaling approaches, such as heterogeneous integration (HI). HI is becoming increasingly implemented in many AI accelerator products due to its potential to enhance overall system performance while also reducing electrical interconnect delays and energy consumption, which are critical for supporting data-intensive AI workloads. In this review, we introduce current and emerging HI technologies and their potential for high-performance systems. We then survey recent industrial and research progress in 3D HI technologies that enable high bandwidth systems and finally, present the emergence of glass core packaging for high-performance AI chip packages.
- 23 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Exploratory Solid-State Computational Devices and Circuits
- Vol. 10, 89-97
Top Cited PapersOpen AccessHeterogeneous Integration Technologies for Artificial Intelligence Applications
The rapid advancement of artificial intelligence (AI) has been enabled by semiconductor-based electronics. However, the conventional methods of transistor scaling are not enough to meet the exponential demand for computing power driven by AI. This has led to a technological shift toward system-level scaling approaches, such as heterogeneous integration (HI). HI is becoming increasingly implemented in many AI accelerator products due to its potential to enhance overall system performance while also reducing electrical interconnect delays and energy consumption, which are critical for supporting data-intensive AI workloads. In this review, we introduce current and emerging HI technologies and their potential for high-performance systems. We then survey recent industrial and research progress in 3D HI technologies that enable high bandwidth systems and finally, present the emergence of glass core packaging for high-performance AI chip packages.
- Open Access
- 17 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of the Electron Devices Society
- Vol. 12, 426-432
Design and Thermal Analysis of 2.5D and 3-D Integrated System of a CMOS Image Sensor and a Sparsity-Aware Accelerator for Autonomous Driving
For the autonomous driving application, data movement has increased rapidly between a CMOS Image sensor (CIS) and the processor due to increase in image resolution. Advanced packaging techniques like 2.5D/3D integration have been proposed to reduce the data movement energy between memory and processor. In this work, we explore the use of such techniques to integrate a CIS and a backend accelerator on a silicon interposer. The data movement energy from CIS to the accelerator is thus reduced by 100× compared to using the conventional MIPI links. We perform thermal simulations to study the impact of the thermal coupling of CIS and accelerator and ensure a peak temperature increase of less than 5 C. We also vary the distance between the CIS and the processor to study the trade-offs between energy savings and peak temperature. Next, we assume the 3D stacked CIS and accelerator to reduce the data movement further and obtain an energy efficiency of 45.81 TOPS/W. Now we observe a heat dissipation challenge with an increase in the peak temperature of more than 85 C. Hence, we scale down the operational frequency and study the trade-off between performance degradation and reduction in peak temperature, while maintaining the accurate multi-object tracking on the BDD100k dataset for autonomous driving.
- 17 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of the Electron Devices Society
- Vol. 12, 426-432
Open AccessDesign and Thermal Analysis of 2.5D and 3-D Integrated System of a CMOS Image Sensor and a Sparsity-Aware Accelerator for Autonomous Driving
For the autonomous driving application, data movement has increased rapidly between a CMOS Image sensor (CIS) and the processor due to increase in image resolution. Advanced packaging techniques like 2.5D/3D integration have been proposed to reduce the data movement energy between memory and processor. In this work, we explore the use of such techniques to integrate a CIS and a backend accelerator on a silicon interposer. The data movement energy from CIS to the accelerator is thus reduced by 100× compared to using the conventional MIPI links. We perform thermal simulations to study the impact of the thermal coupling of CIS and accelerator and ensure a peak temperature increase of less than 5 C. We also vary the distance between the CIS and the processor to study the trade-offs between energy savings and peak temperature. Next, we assume the 3D stacked CIS and accelerator to reduce the data movement further and obtain an energy efficiency of 45.81 TOPS/W. Now we observe a heat dissipation challenge with an increase in the peak temperature of more than 85 C. Hence, we scale down the operational frequency and study the trade-off between performance degradation and reduction in peak temperature, while maintaining the accurate multi-object tracking on the BDD100k dataset for autonomous driving.
- 8 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 31 (10) , 1592-1602
H3DAtten: Heterogeneous 3-D Integrated Hybrid Analog and Digital Compute-in-Memory Accelerator for Vision Transformer Self-Attention
After the success of the transformer networks on natural language processing (NLP), the application of transformers to computer vision (CV) has followed suit to deliver unprecedented performance gains on vision tasks, including image recognition and object detection. The multihead self-attention (MHSA) is the key component in transformers, allowing the models to learn the amount of attention paid to each input position. Despite its strong modeling capability, MHSA involves complex operations that make transformers prohibitively costly for hardware deployment. Existing acceleration efforts with conventional hardware platforms are challenged by the memory wall. To alleviate the memory wall problem, compute-in-memory (CIM) is a promising solution by storing all model parameters on-chip in compute-capable memory arrays. The footprint of 2-D CIM designs must, however, expand to accommodate the increasingly larger model sizes. In this work, we present a heterogeneous 3-D integrated (H3D) accelerator to target the MHSA workloads in vision transformers. H3D allows the proposed H3DAtten architecture to combine the merits of resistive random access memory (RRAM)-based analog CIM (ACIM) in 40 nm and static random access memory (SRAM)-based digital CIM (DCIM) in 16 nm. We perform comprehensive signaling and thermal analyses to examine the effects of 3-D stacking on the accelerator. Compared to iso-capacity 2-D baseline designs, the proposed 5-tier H3DAtten accelerator achieves 8.4 $\times$ compute density without experiencing accuracy loss on the ImageNet-1k dataset.
- 8 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 31 (10) , 1592-1602
H3DAtten: Heterogeneous 3-D Integrated Hybrid Analog and Digital Compute-in-Memory Accelerator for Vision Transformer Self-Attention
After the success of the transformer networks on natural language processing (NLP), the application of transformers to computer vision (CV) has followed suit to deliver unprecedented performance gains on vision tasks, including image recognition and object detection. The multihead self-attention (MHSA) is the key component in transformers, allowing the models to learn the amount of attention paid to each input position. Despite its strong modeling capability, MHSA involves complex operations that make transformers prohibitively costly for hardware deployment. Existing acceleration efforts with conventional hardware platforms are challenged by the memory wall. To alleviate the memory wall problem, compute-in-memory (CIM) is a promising solution by storing all model parameters on-chip in compute-capable memory arrays. The footprint of 2-D CIM designs must, however, expand to accommodate the increasingly larger model sizes. In this work, we present a heterogeneous 3-D integrated (H3D) accelerator to target the MHSA workloads in vision transformers. H3D allows the proposed H3DAtten architecture to combine the merits of resistive random access memory (RRAM)-based analog CIM (ACIM) in 40 nm and static random access memory (SRAM)-based digital CIM (DCIM) in 16 nm. We perform comprehensive signaling and thermal analyses to examine the effects of 3-D stacking on the accelerator. Compared to iso-capacity 2-D baseline designs, the proposed 5-tier H3DAtten accelerator achieves 8.4 $\times$ compute density without experiencing accuracy loss on the ImageNet-1k dataset.
- 29 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 70 (2) , 485-492
3-D Heterogeneous Integration of RRAM-Based Compute-In-Memory: Impact of Integration Parameters on Inference Accuracy
Three-dimensional heterogeneous integration (3-D-HI) has been proposed as a potential method to stack a large amount of embedded memory required in state-of-the-art compute-in-memory (CIM) artificial intelligence (AI) accelerators. While embedded nonvolatile memory, such as resistive RAM (RRAM), is a promising alternative to static random access memory (SRAM)/dynamic random access memory (DRAM) as a CIM synaptic device owing to high density, low leakage, and nondestructive read, thermal-induced device conductance drift remains a challenge. High-temperature-driven lower retention can be more significant in dense memory-logic 3-D integration due to increased volumetric power, which has not been studied in prior work. The scope of this work is to quantify the thermal impact of different 3-D-HI architectures on the reliability of 3-D-integrated binary RRAM devices for CIM applications. A device-integration-application reliability evaluation methodology is proposed, using which 3-D integration architectures and logic-memory partitioning configurations are benchmarked. Due to higher junction temperatures for memory tier in both five-tier monolithic 3-D (M3D) and five-tier through silicon via (TSV)-based 3-D compared to the 2-D baseline, the drop in inference accuracy at ten years is $\approx$ 80%. For our assumed device, integration, and application parameters, a three-tier configuration provides a balanced design option between thermal and application performance. The integrated benchmark framework is released on GitHub (https://github.com/i3dsystems/3D_CIM_thermal_v1.0) as an open-source tool for the research community.
- 29 December 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 70 (2) , 485-492
3-D Heterogeneous Integration of RRAM-Based Compute-In-Memory: Impact of Integration Parameters on Inference Accuracy
Three-dimensional heterogeneous integration (3-D-HI) has been proposed as a potential method to stack a large amount of embedded memory required in state-of-the-art compute-in-memory (CIM) artificial intelligence (AI) accelerators. While embedded nonvolatile memory, such as resistive RAM (RRAM), is a promising alternative to static random access memory (SRAM)/dynamic random access memory (DRAM) as a CIM synaptic device owing to high density, low leakage, and nondestructive read, thermal-induced device conductance drift remains a challenge. High-temperature-driven lower retention can be more significant in dense memory-logic 3-D integration due to increased volumetric power, which has not been studied in prior work. The scope of this work is to quantify the thermal impact of different 3-D-HI architectures on the reliability of 3-D-integrated binary RRAM devices for CIM applications. A device-integration-application reliability evaluation methodology is proposed, using which 3-D integration architectures and logic-memory partitioning configurations are benchmarked. Due to higher junction temperatures for memory tier in both five-tier monolithic 3-D (M3D) and five-tier through silicon via (TSV)-based 3-D compared to the 2-D baseline, the drop in inference accuracy at ten years is $\approx$ 80%. For our assumed device, integration, and application parameters, a three-tier configuration provides a balanced design option between thermal and application performance. The integrated benchmark framework is released on GitHub (https://github.com/i3dsystems/3D_CIM_thermal_v1.0) as an open-source tool for the research community.
- 21 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Components, Packaging and Manufacturing Technology
- Vol. 12 (11) , 1824-1831
Analysis of Power Delivery Network (PDN) in Bridge-Chips for 2.5-D Heterogeneous Integration
In this paper, we investigate the impact of power delivery network (PDN) in bridge-chip based 2.5-D/3-D heterogeneous integration platforms. The focus of the paper is bridge-chip based CPU-FPGA and FPGA-Stacked Memory integration technologies. While bridge-chip based interconnect platforms present PDN challenges, depending on the power map, including a PDN in the bridge-chip can help reduce the impact significantly. We perform three case studies: (1) inclusion of ground network in the bridge-chip, (2) inclusion of power and ground network in the bridge-chip, and (3) inclusion of metal-insulator-metal (MIM) decoupling capacitors in the bridge-chip. Inclusion of both power and ground network can reduce DC IR drop by ~20% for a CPU-FPGA integration case study and by ~40% for an FPGA-stacked memory configuration. Our L di/dt noise analysis shows that if we include decoupling capacitors in the bridge-chip, we can significantly reduce the high frequency ripple in the power supply. We also perform a design space exploration for power delivery with the following parameters: resistance of the PDN in the bridge-chip, decoupling capacitor density in the bridge-chip, and overlap region between a bridge-chip and active dice.
- 21 November 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Components, Packaging and Manufacturing Technology
- Vol. 12 (11) , 1824-1831
Analysis of Power Delivery Network (PDN) in Bridge-Chips for 2.5-D Heterogeneous Integration
In this paper, we investigate the impact of power delivery network (PDN) in bridge-chip based 2.5-D/3-D heterogeneous integration platforms. The focus of the paper is bridge-chip based CPU-FPGA and FPGA-Stacked Memory integration technologies. While bridge-chip based interconnect platforms present PDN challenges, depending on the power map, including a PDN in the bridge-chip can help reduce the impact significantly. We perform three case studies: (1) inclusion of ground network in the bridge-chip, (2) inclusion of power and ground network in the bridge-chip, and (3) inclusion of metal-insulator-metal (MIM) decoupling capacitors in the bridge-chip. Inclusion of both power and ground network can reduce DC IR drop by ~20% for a CPU-FPGA integration case study and by ~40% for an FPGA-stacked memory configuration. Our L di/dt noise analysis shows that if we include decoupling capacitors in the bridge-chip, we can significantly reduce the high frequency ripple in the power supply. We also perform a design space exploration for power delivery with the following parameters: resistance of the PDN in the bridge-chip, decoupling capacitor density in the bridge-chip, and overlap region between a bridge-chip and active dice.
- 23 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 12 (2) , 445-457
A Compute-in-Memory Hardware Accelerator Design With Back-End-of-Line (BEOL) Transistor Based Reconfigurable Interconnect
Compute-in-memory (CIM) paradigm using ferroelectric field effect transistor (FeFET) as the weight element is projected to exhibit excellent energy efficiency for accelerating deep neural network (DNN) inference. However, two challenges exist. On the technology level, the chip area scaling is stalled due to the lack of logic voltage compatible FeFET at leading-edge technology node, e. g. 7nm. On the system level, CIM-based inference engine designs are usually customized for a specific DNN model, lacking the flexibility to support different DNN models. Besides, communication latency varies across different DNN models and can bound the total inference latency. Therefore, a reconfigurable interconnect is desired to be adaptive to different workloads, which can induce high area cost due to the reconfigurable circuit modules. To solve these issues, in this work, a system-technology co-design (STCO) of a monolithic 3D (M3D) reconfigurable CIM accelerator is performed, where back-end-of-line (BEOL) compatible oxide channel MOSFET and FeFET technologies are utilized. On the technology level, W-doped indium oxide (IWO) NMOS is utilized to design area-efficient M3D write circuit. On the system level, a reconfigurable interconnect design that inserts workload-specific express link is proposed, where the IWO-based NMOS and FeFET are adopted as the building element of the mux and crossbar switch in the router. The algorithm for interconnect configuration is also devised to achieve optimal latency for different workloads. From the system-level evaluation results, M3D IWO FeFET design (utilizing a hybrid 22nm/7nm M3D partition) shows $3.1\times $ times higher energy efficiency than a 7nm 2D SRAM design with comparable chip area. With the proposed reconfigurable interconnect scheme, the interconnect latency is reduced by 9%~32% compared to the baseline with a regular mesh network.
- 23 May 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
- Vol. 12 (2) , 445-457
A Compute-in-Memory Hardware Accelerator Design With Back-End-of-Line (BEOL) Transistor Based Reconfigurable Interconnect
Compute-in-memory (CIM) paradigm using ferroelectric field effect transistor (FeFET) as the weight element is projected to exhibit excellent energy efficiency for accelerating deep neural network (DNN) inference. However, two challenges exist. On the technology level, the chip area scaling is stalled due to the lack of logic voltage compatible FeFET at leading-edge technology node, e. g. 7nm. On the system level, CIM-based inference engine designs are usually customized for a specific DNN model, lacking the flexibility to support different DNN models. Besides, communication latency varies across different DNN models and can bound the total inference latency. Therefore, a reconfigurable interconnect is desired to be adaptive to different workloads, which can induce high area cost due to the reconfigurable circuit modules. To solve these issues, in this work, a system-technology co-design (STCO) of a monolithic 3D (M3D) reconfigurable CIM accelerator is performed, where back-end-of-line (BEOL) compatible oxide channel MOSFET and FeFET technologies are utilized. On the technology level, W-doped indium oxide (IWO) NMOS is utilized to design area-efficient M3D write circuit. On the system level, a reconfigurable interconnect design that inserts workload-specific express link is proposed, where the IWO-based NMOS and FeFET are adopted as the building element of the mux and crossbar switch in the router. The algorithm for interconnect configuration is also devised to achieve optimal latency for different workloads. From the system-level evaluation results, M3D IWO FeFET design (utilizing a hybrid 22nm/7nm M3D partition) shows $3.1\times $ times higher energy efficiency than a 7nm 2D SRAM design with comparable chip area. With the proposed reconfigurable interconnect scheme, the interconnect latency is reduced by 9%~32% compared to the baseline with a regular mesh network.
- 23 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 68 (11) , 5598-5605
Heterogeneous 3-D Integration of Multitier Compute-in-Memory Accelerators: An Electrical-Thermal Co-Design
Emerging nonvolatile memory (eNVM)-based compute-in-memory (CIM) accelerators have been proven in silicon for machine learning at the macrolevel. To fully unleash the system-level benefits, the heterogeneous 3-D integration (H3D) using through-silicon via (TSV) is a promising approach, to: 1) address the challenges of area-hungry peripheries in CIM accelerators; 2) solve the 2-D scaling challenges of eNVM; and 3) stack enormous amount of embedded memories that are required in state-of-the-art deep neural network models. This article presents an electrical-thermal co-design of multitier CIM accelerators, based on SRAM and/or eNVM, with hybrid technology nodes for logic and memory tiers. We benchmark the CIM accelerators on 8-bit ResNet-34 for ImageNet recognition, with layer-by-layer and pipelined schemes, respectively. By sweeping TSV diameter from $30~\mu \text{m}$ to 100 nm, we investigate the tradeoffs of system performance metrics (TOPS/W, TOPS, and TOPS/mm 2 ) and H3D challenges (thermal and IR-drop in power delivery). Finally, we find the sweet spot of TSV diameter for multitier H3D system is 1– $3~\mu \text{m}$ , to guarantee balanced area-overhead, performance, and IR-drop in power delivery. The extended benchmark framework is released on GitHub ( https://github.com/neurosim ) as an open-source tool for the research community.
- 23 September 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Electron Devices
- Vol. 68 (11) , 5598-5605
Heterogeneous 3-D Integration of Multitier Compute-in-Memory Accelerators: An Electrical-Thermal Co-Design
Emerging nonvolatile memory (eNVM)-based compute-in-memory (CIM) accelerators have been proven in silicon for machine learning at the macrolevel. To fully unleash the system-level benefits, the heterogeneous 3-D integration (H3D) using through-silicon via (TSV) is a promising approach, to: 1) address the challenges of area-hungry peripheries in CIM accelerators; 2) solve the 2-D scaling challenges of eNVM; and 3) stack enormous amount of embedded memories that are required in state-of-the-art deep neural network models. This article presents an electrical-thermal co-design of multitier CIM accelerators, based on SRAM and/or eNVM, with hybrid technology nodes for logic and memory tiers. We benchmark the CIM accelerators on 8-bit ResNet-34 for ImageNet recognition, with layer-by-layer and pipelined schemes, respectively. By sweeping TSV diameter from $30~\mu \text{m}$ to 100 nm, we investigate the tradeoffs of system performance metrics (TOPS/W, TOPS, and TOPS/mm 2 ) and H3D challenges (thermal and IR-drop in power delivery). Finally, we find the sweet spot of TSV diameter for multitier H3D system is 1– $3~\mu \text{m}$ , to guarantee balanced area-overhead, performance, and IR-drop in power delivery. The extended benchmark framework is released on GitHub ( https://github.com/neurosim ) as an open-source tool for the research community.
- 1 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1459-1467
BEOL-Embedded 3D Polylithic Integration: Thermal and Interconnection Considerations
In this paper, a BEOL-embedded chiplet integration scheme is proposed for dense 3D integration. This scheme represents a system with multiple device tiers (primary and embedded tiers) where custom chiplets, such as voltage regulator modules, I/O drivers, RF front-end chips, etc. are embedded into the BEOL of an application processor tier with a monolithic memory device tier, such as resistive RAM. The proposed integration scheme is thermally evaluated to investigate the effects of design parameters on the thermal operation of the primary and embedded tiers, along with transient analysis to estimate the extent of inter-tier thermal coupling. Results for steady-state operation suggest the thermal viability of such an integration approach. Considerable reduction in inter-tier thermal coupling for transient operation was achieved with dual-sided cooling (tier 1 to tier 3 coupling: 85% for air-cooling, reduced to 65% for dual-sided cooling). We also propose and demonstrate metal electroless plating in addition to mechanical self-alignment as an enabling technology for BEOL-embedded integration to facilitate low-temperature, low-pressure, and high interconnect density inter-die bonding. Post assembly Ni electroless deposited bonding was verified across all inspected diagonal pillars in an area array of 50×50 copper pillars.
- 1 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1459-1467
BEOL-Embedded 3D Polylithic Integration: Thermal and Interconnection Considerations
In this paper, a BEOL-embedded chiplet integration scheme is proposed for dense 3D integration. This scheme represents a system with multiple device tiers (primary and embedded tiers) where custom chiplets, such as voltage regulator modules, I/O drivers, RF front-end chips, etc. are embedded into the BEOL of an application processor tier with a monolithic memory device tier, such as resistive RAM. The proposed integration scheme is thermally evaluated to investigate the effects of design parameters on the thermal operation of the primary and embedded tiers, along with transient analysis to estimate the extent of inter-tier thermal coupling. Results for steady-state operation suggest the thermal viability of such an integration approach. Considerable reduction in inter-tier thermal coupling for transient operation was achieved with dual-sided cooling (tier 1 to tier 3 coupling: 85% for air-cooling, reduced to 65% for dual-sided cooling). We also propose and demonstrate metal electroless plating in addition to mechanical self-alignment as an enabling technology for BEOL-embedded integration to facilitate low-temperature, low-pressure, and high interconnect density inter-die bonding. Post assembly Ni electroless deposited bonding was verified across all inspected diagonal pillars in an area array of 50×50 copper pillars.
- 22 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Components, Packaging and Manufacturing Technology
- Vol. 9 (12) , 2393-2403
Microfluidic Cooling of a 14-nm 2.5-D FPGA With 3-D Printed Manifolds for High-Density Computing: Design Considerations, Fabrication, and Electrical Characterization
2.5D integration is becoming a common method of tightly integrating heterogeneous dice with dense interconnects for efficient, high-bandwidth inter-die communication. While this tight integration improves performance, it also increases the challenge of heat extraction by increasing aggregate package powers and introducing thermal cross-talk between adjacent dice. In this work, a microfluidic heat sink is used to cool a 2.5D Stratix 10 GX FPGA consisting of an FGPA die surrounded by four transceiver dice. The heat sink utilizes a heterogeneous micropin-fin array with micropin-fin densities which are tailored to the local heat fluxes of the underlying dice. Enabled through a 3D-printed enclosure for fluid delivery, the assembled heat sink has a total height of 6.5, including the tubes used for fluid delivery. The heat sink is tested in an open loop system with deionized water as a coolant and thermal performance is compared against a high-end air cooled heat sink. Improvements in die temperatures, computational density, and thermal coupling between dice are observed. The effect of the FPGA power on the surrounding transceiver die temperatures was reduced by a factor of 10× to over 100× when compared with the air cooled heat sink.
- 22 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Components, Packaging and Manufacturing Technology
- Vol. 9 (12) , 2393-2403
Microfluidic Cooling of a 14-nm 2.5-D FPGA With 3-D Printed Manifolds for High-Density Computing: Design Considerations, Fabrication, and Electrical Characterization
2.5D integration is becoming a common method of tightly integrating heterogeneous dice with dense interconnects for efficient, high-bandwidth inter-die communication. While this tight integration improves performance, it also increases the challenge of heat extraction by increasing aggregate package powers and introducing thermal cross-talk between adjacent dice. In this work, a microfluidic heat sink is used to cool a 2.5D Stratix 10 GX FPGA consisting of an FGPA die surrounded by four transceiver dice. The heat sink utilizes a heterogeneous micropin-fin array with micropin-fin densities which are tailored to the local heat fluxes of the underlying dice. Enabled through a 3D-printed enclosure for fluid delivery, the assembled heat sink has a total height of 6.5, including the tubes used for fluid delivery. The heat sink is tested in an open loop system with deionized water as a coolant and thermal performance is compared against a high-end air cooled heat sink. Improvements in die temperatures, computational density, and thermal coupling between dice are observed. The effect of the FPGA power on the surrounding transceiver die temperatures was reduced by a factor of 10× to over 100× when compared with the air cooled heat sink.