Editorial Material
Computer Science, Software Engineering
Yufei Ding, Ali Javadi-Abhari
Summary: This special issue presents four articles that discuss the technical challenges and opportunities of quantum computing. They cover topics such as mitigating decoherence errors, emerging applications, software system design guidelines, and hardware error characterization in the quantum cloud. These articles have the potential to enhance the infrastructure and benefit the users of quantum cloud services.
IEEE INTERNET COMPUTING
(2022)
Article
Computer Science, Information Systems
Ping-Jing Lu, Ming-Che Lai, Jun-Sheng Chang
Summary: This article introduces the importance of high-performance interconnection networks in high-performance computing systems, as well as the trends and technologies in the development of high-performance computing. It also analyzes the main interconnection networks currently used and discusses the future challenges and prospects.
Article
Materials Science, Multidisciplinary
Xiaoyue Wang, Chi Liu, Yuning Wei, Shun Feng, Dongming Sun, Huiming Cheng
Summary: Since the 1960s, the size of metal oxide semiconductor field-effect transistors has been scaled down to increase transistor density on a chip. New materials and technologies have been introduced, including SiGe strained silicon and high-k metal gate. The development of devices using 3D structures and low-dimensional materials is reviewed, as they are expected to continue the Moore's law further.
Article
Business
Christophe Lecuyer
Summary: Moore's Law has played a crucial role in driving process and product innovation in the semiconductor industry, and it significantly contributed to Intel's dominance and competitiveness.
ENTERPRISE & SOCIETY
(2022)
Article
Chemistry, Analytical
Chun Fei Siah, Lucas Yu Xiang Lum, Jianxiong Wang, Simon Chun Kiat Goh, Chong Wei Tan, Liangxing Hu, Philippe Coquet, Hong Li, Chuan Seng Tan, Beng Kang Tay
Summary: A CMOS-compatible CNT array transfer method has been developed to address issues preventing the widespread integration of CNTs onto device applications. Using an alignment tool with thermocompression capabilities enables highly accurate transfer of CNT arrays, with pointers for further improvement in transferred CNT quality discussed.
Review
Physics, Applied
Dario L. Goldfarb
Summary: This paper reviews the use of lithographic materials in semiconductor patterning applications and discusses different types of photoresist platforms. It also explores the shift from Moore's Law 1.0, which focused on increasing the number of components on a chip, to Moore's Law 2.0, which emphasizes 3D scaling and integration of new technologies such as AI and 3D packaging.
JAPANESE JOURNAL OF APPLIED PHYSICS
(2022)
Article
Computer Science, Hardware & Architecture
Seyed Erfan Fatemieh, Samira Shirinabadi Farahani, Mohammad Reza Reshadinezhad
Summary: This article presents a low power, area-efficient full adder cell designed with approximate outputs, showing up to 72% improvement in energy efficiency. The proposed cell is evaluated in four different scenarios, demonstrating its power consumption and performance under various conditions. Additionally, MATLAB is used to assess the proposed cell in image processing applications, with acceptable results based on image quality metrics like PSNR.
SUSTAINABLE COMPUTING-INFORMATICS & SYSTEMS
(2021)
Article
Computer Science, Information Systems
Yinqiu Liu, Kai Qian, Kun Wang, Lei He
Summary: This article evaluates various blockchain proposals over the past decade and finds the potential for consensus-based scaling to be limited. It introduces topology and hardware innovations for scaling blockchain.
IEEE SYSTEMS JOURNAL
(2022)
Article
Materials Science, Multidisciplinary
Manuel Schroter, Congwei Wang, Marco Terrigno, Philipp Hornauer, Ziqiang Huang, Ravi Jagasia, Andreas Hierlemann
Summary: This study utilizes high-density microelectrode arrays to perform large-scale electrophysiological recordings on human cerebral organoids, providing insights into their functionality and potential functional connectivity, leading to a better understanding of developing neuronal networks in brain organoids.
Article
Nanoscience & Nanotechnology
Thomas Palmer, Erik H. Waller, Heiko Andra, Konrad Steiner, Georg von Freymann
Summary: The study investigates the impact of light-particle interactions and varying scanning speed on heat generation and particle deposition in direct laser writing of silver. A theoretical model is introduced which captures diffusion of particles and heat as well as the fluid dynamics of the photoresist. The simulations suggest optical forces generally surmount those due to convection of the photoresist.
ACS APPLIED NANO MATERIALS
(2021)
Article
Chemistry, Multidisciplinary
Fei Hui, Conghui Zhang, Huanhuan Yu, Tingting Han, Jonas Weber, Yaqing Shen, Yiping Xiao, Xiaohong Li, Zhijun Zhang, Peisong Liu
Summary: Janus 2D materials have attracted considerable attention due to their exceptional interface properties and potential applications in optoelectronic devices. However, the fabrication of high-quality Janus 2D materials at a scalable level remains a major challenge. In this study, a new method called chemical breakdown is developed to produce large-area uniform Janus graphene oxide (J-GO) films with high-quality. Furthermore, the first application of J-GO in memristive devices for neuromorphic computing is demonstrated, showing enhanced performance in terms of resistive switching, low leakage current, low operation voltage, high endurance, and electro-synaptic plasticity.
ADVANCED FUNCTIONAL MATERIALS
(2023)
Article
Environmental Sciences
Pasquale Imperatore, Antonio Pepe, Eugenio Sansosti
Summary: This paper presents a perspective on the state-of-the-art of high performance computing (HPC) methodologies applied to spaceborne SAR interferometry, discussing different parallel algorithms for interferometric processing of SAR data at various levels. It emphasizes on key processing steps and computational relevance, examines existing implementations of InSAR stages using diverse parallel strategies and architectures, and analyzes selected InSAR computational schemes in the literature to outline potentialities and limitations. The survey focuses on inherent computational approaches enabling large-scale interferometric SAR processing, offering insight into open issues and outlining future trends in the field.
Article
Computer Science, Hardware & Architecture
Jeff Anderson, Engin Kayraklioglu, Hamid Reza Imani, Chen Shen, Mario Miscuglio, Volker J. Sorger, Tarek El-Ghazawi
Summary: Innovative processor architectures, such as the Reconfigurable Optical Computer (ROC), are designed to sustain performance improvements in the post-Moore's Law era. ROC solves partial differential equations in one shot using a mesh of optical components, leveraging the advantages of optical computing. Lightweight virtualization is introduced to address the limitations of the hardware, and various resource scheduling strategies are investigated, resulting in a 2x speedup with minimal overhead and errors under 10%.
ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS
(2023)
Article
Computer Science, Information Systems
Tommaso Rizzo, Sebastiano Strangio, Giuseppe Iannaccone
Summary: Improving the energy efficiency of deep learning systems is crucial for edge devices and data centers. By utilizing analog vector-matrix multipliers, the energy consumption per operation can be significantly reduced. We present a design of an analog VMM using industry-standard technology, which achieves high performance in terms of energy efficiency.
Article
Chemistry, Multidisciplinary
Michael T. Jones, Md Serajum Monir, Felix N. Krauth, Pascal Macha, Yu-Ling Hsueh, Angus Worrall, Joris G. Keizer, Ludwik Kranz, Samuel K. Gorman, Yousun Chung, Rajib Rahman, Michelle Y. Simmons
Summary: This study demonstrates a platform for quantum computing using phosphorus donor atoms in silicon. By adjusting the arrangement and spacing of the donor atoms, the speed and accuracy of qubit gates can be improved without affecting neighboring qubits. This approach provides a new pathway for achieving high fidelity and scalable quantum computing.
Article
Computer Science, Hardware & Architecture
George Michelogiannakis, Benjamin Klenk, Brandon Cook, Min Yee Teh, Madeleine Glick, Larry Dennison, Keren Bergman, John Shalf
Summary: With the emergence of specialized accelerators, there is a need for more flexible resource allocation in HPC systems. This study analyzes NERSC's Cori system and profiles deep-learning applications to investigate hardware resource disaggregation. The results demonstrate that intra-rack disaggregation can effectively locate required resources and reduce the utilization of memory and NICs.
ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION
(2022)
Article
Computer Science, Hardware & Architecture
Chao Zhang, Maximilian Bremer, C. Y. Chan, John Shalf, Xiaochen Guo
Summary: Sparse linear algebra is crucial in many applications. The column-wise SpGEMM algorithm has good locality but the sparse accumulation (SPA) step remains a performance bottleneck. This work proposes the ASA architecture to overcome the challenges of SPA and achieves significant speedup.
ACM TRANSACTIONS ON ARCHITECTURE AND CODE OPTIMIZATION
(2022)
Article
Computer Science, Interdisciplinary Applications
Erik W. Draeger, Andrew Siegel
Summary: This article discusses the historical progress of supercomputing capabilities and the challenges of achieving exascale systems, as well as providing insights on designing and implementing scientific applications in the exascale and post-exascale eras.
COMPUTING IN SCIENCE & ENGINEERING
(2023)
Article
Engineering, Electrical & Electronic
Zhenguo Wu, Liang Yuan Dai, Asher Novick, Madeleine Glick, Ziyi Zhu, Sebastien Rumley, George Michelogiannakis, John Shalf, Keren Bergman
Summary: SiPAC is a Silicon Photonic Accelerated Compute cluster that accelerates distributed deep learning training by utilizing embedded photonics and a novel collective algorithm. It expedites collective communications commonly used in deep learning, effectively solving the communication bottleneck problem.
JOURNAL OF LIGHTWAVE TECHNOLOGY
(2023)
Proceedings Paper
Computer Science, Hardware & Architecture
George Michelogiannakis, Yehia Arafa, Brandon Cook, Liang Yuan Dai, Abdel-Hameed (Hameed) Badawy, Madeleine Glick, Yuyang Wang, Keren Bergman, John Shalf
Summary: Resource disaggregation is a promising approach to meet the diverse workload requirements of HPC systems. By co-designing modern photonics with HPC racks, flexible intra-rack resource disaggregation can be implemented, effectively meeting the bandwidth and latency requirements of HPC applications.
2023 IEEE INTERNATIONAL CONFERENCE ON CLUSTER COMPUTING, CLUSTER
(2023)
Article
Computer Science, Interdisciplinary Applications
Satoshi Matsuoka, Jens Domke, Mohamed Wahib, Aleksandr Drozd, Andrew A. Chien, Raymond Bair, Jeffrey S. Vetter, John Shalf
Summary: The translation discusses the importance of designing proxy applications for co-designing and evaluating current generation supercomputers, and highlights the need to adapt to future heterogeneity.
COMPUTING IN SCIENCE & ENGINEERING
(2022)
Proceedings Paper
Computer Science, Hardware & Architecture
Md Abdul M. Faysal, Shaikh Arifuzzaman, Cy Chan, Maximilian Bremer, Doru Popovici, John Shalf
Summary: Community detection is crucial in graph analysis, and using an information-theoretic approach like Infomap can deliver better solutions. However, the inherently sequential nature of Infomap algorithm limits its scalability for large networks. This work introduces a hybrid parallel approach, HyPC-Map, which achieves a significant speedup of 25 times without sacrificing solution quality.
2021 IEEE HIGH PERFORMANCE EXTREME COMPUTING CONFERENCE (HPEC)
(2021)
Proceedings Paper
Computer Science, Software Engineering
Tan Nguyen, Erich Strohmaier, John Shalf
Summary: Automating the workload characterization process in hardware design is crucial, especially with the increasing complexity of applications and input problems. A new approach based on code similarity learning is proposed, decomposing applications into small kernels mapped to known patterns for reusing behavior on hardware. This approach involves a new code representation and similarity metric, automated detection using compiler and machine learning methods, and distinguishing kernels in benchmarks for suggesting optimal configurations for hardware accelerators.
PROCEEDINGS OF THE SEVENTH ANNUAL WORKSHOP ON THE LLVM COMPILER INFRASTRUCTURE IN HPC (LLVM-HPC2021)
(2021)
Proceedings Paper
Engineering, Electrical & Electronic
George Michelogiannakis, Min Yeh Teh, Madeleine Glick, John Shalf, Keren Bergman
Summary: Photonic switches have the potential to have a significant impact on future HPC and datacenter networks, but system-wide benefits cannot be achieved without changing other parts of the system architecture. Therefore, co-design between photonic switches and the rest of the system is necessary to adapt the system network and support system-wide trends.
OPTICAL INTERCONNECTS XXI
(2021)
Article
Computer Science, Hardware & Architecture
Georgios Tzimpragos, Jennifer Volk, Dilip Vasudevan, Nestan Tsiskaridze, George Michelogiannakis, Advait Madhavan, John Shalf, Timothy Sherwood
Summary: As we approach the end of CMOS transistor scaling, it is necessary to explore novel computing methods to meet the growing demands. The intersection of superconductor electronics and delay-coded computation could provide a more efficient foundation for computing.
Article
Computer Science, Interdisciplinary Applications
John West
Summary: U.S. computing leaders are collaborating with various entities to accelerate research on responses to COVID-19, providing resources such as some of the world's fastest computers. Efforts include understanding how the virus spreads, its biology and structure, as well as developing vaccines and antiviral treatments. The rapid mobilization of national research cyberinfrastructure highlights the importance of ongoing investments in large-scale scientific computing.
COMPUTING IN SCIENCE & ENGINEERING
(2021)
Proceedings Paper
Computer Science, Information Systems
Min Yee Teh, Yu-Han Hung, George Michelogiannakis, Shijia Yan, Madeleine Glick, John Shalf, Keren Bergman
PROCEEDINGS OF SC20: THE INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS (SC20)
(2020)
Proceedings Paper
Computer Science, Hardware & Architecture
Nicholas Buoncristiani, Sanjana Shah, David Donofrio, John Shalf
2020 IEEE 34TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM IPDPS 2020
(2020)
Proceedings Paper
Computer Science, Theory & Methods
Muhammad Nufail Farooqi, Tan Nguyen, Weiqun Zhang, Ann S. Almgren, John Shalf, Didem Unat
HIGH PERFORMANCE COMPUTING: ISC HIGH PERFORMANCE 2019 INTERNATIONAL WORKSHOPS
(2020)