Shuichi Sakai
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
Shuichi Sakai has a total of 53 co-authors. Here are the top co-authors by number of shared publications:
- Hidetsugu Irie
The University of Tokyo, Tokyo, Japan
18 shared publications - Junichiro Kadomoto
The University of Tokyo, Tokyo, Japan
13 shared publications - Ryota Shioya
The University of Tokyo
7 shared publications - Toru Koizumi
PEZY Computing, K.K., Tokyo, Japan
4 shared publications - Yuya Degawa
Fujitsu Limited, Kanagawa, Japan
4 shared publications - Shu Sugita
The University of Tokyo, Tokyo, Japan
3 shared publications
Latest Publications
- Open Access
- 26 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 25 (1) , 17-20
Inspex: Speculative Execution of Ready-to-Execute Loads in In-Order Cores
In-order (InO) cores are processor cores that execute instructions in program order. Due to their low complexity, InO cores have been widely used in situations where energy efficiency and small circuit area are required, but they provide limited performance. We focus on stalls in InO cores that are caused by load instructions and their consumer instructions. These stalls significantly degrade the performance, which accounts for 70% of the total execution time on SPEC CPU 2017. We found that many of these load instructions are ready to execute and could have been issued earlier. Based on this observation, we propose Inspex, which improves the performance of InO cores while maintaining its simplicity. Inspex predicts ready-to-execute loads and speculatively pre-executes them. This enables obtaining load results earlier, thereby removing the stalls caused by load consumers. Our simulation results show that Inspex improved the performance by 25.6% and reduced the energy consumption by 10.6%, and reduced the energy-delay product by 28.0% compared with a baseline InO core while incurring an area overhead of 0.65%.
- 26 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 25 (1) , 17-20
Open AccessInspex: Speculative Execution of Ready-to-Execute Loads in In-Order Cores
In-order (InO) cores are processor cores that execute instructions in program order. Due to their low complexity, InO cores have been widely used in situations where energy efficiency and small circuit area are required, but they provide limited performance. We focus on stalls in InO cores that are caused by load instructions and their consumer instructions. These stalls significantly degrade the performance, which accounts for 70% of the total execution time on SPEC CPU 2017. We found that many of these load instructions are ready to execute and could have been issued earlier. Based on this observation, we propose Inspex, which improves the performance of InO cores while maintaining its simplicity. Inspex predicts ready-to-execute loads and speculatively pre-executes them. This enables obtaining load results earlier, thereby removing the stalls caused by load consumers. Our simulation results show that Inspex improved the performance by 25.6% and reduced the energy consumption by 10.6%, and reduced the energy-delay product by 28.0% compared with a baseline InO core while incurring an area overhead of 0.65%.
- Open Access
- 25 February 2025
- Published by Association for Computing Machinery (ACM)
- p. 167-179
Biotite: A High-Performance Static Binary Translator using Source-Level Information
Research on novel Instruction Set Architectures (ISAs) is actively pursued; however, it requires extensive efforts to develop and maintain comprehensive compilation toolchains for each new ISA. Binary translation can provide a practical solution for ISA researchers to port target programs to novel ISAs when such a level of toolchain support is not available. However, to ensure the correct handling of indirect jumps, existing binary translators rely on complex runtime systems, whose implementation on primitive research ISAs demands significant efforts. ISA researchers generally have access to additional source-level information, including the symbol table and source code, when using binary translators. The symbol table can provide potential jump targets for optimizing indirect jumps, and ISA-independent functions in source code can be directly compiled without translation. Leveraging source-level information as additional input, in this paper, we propose Biotite, a high-performance static binary translator that correctly handles arbitrary indirect jumps. Currently, Biotite supports the translation of RV64GC Linux binaries to self-contained LLVM IR. Our evaluation shows that Biotite successfully translates all benchmarks in SPEC CPU 2017 and achieves a 2.346× performance improvement over QEMU for the integer benchmark suite.
- 25 February 2025
- Published by Association for Computing Machinery (ACM)
- p. 167-179
Open AccessBiotite: A High-Performance Static Binary Translator using Source-Level Information
Research on novel Instruction Set Architectures (ISAs) is actively pursued; however, it requires extensive efforts to develop and maintain comprehensive compilation toolchains for each new ISA. Binary translation can provide a practical solution for ISA researchers to port target programs to novel ISAs when such a level of toolchain support is not available. However, to ensure the correct handling of indirect jumps, existing binary translators rely on complex runtime systems, whose implementation on primitive research ISAs demands significant efforts. ISA researchers generally have access to additional source-level information, including the symbol table and source code, when using binary translators. The symbol table can provide potential jump targets for optimizing indirect jumps, and ISA-independent functions in source code can be directly compiled without translation. Leveraging source-level information as additional input, in this paper, we propose Biotite, a high-performance static binary translator that correctly handles arbitrary indirect jumps. Currently, Biotite supports the translation of RV64GC Linux binaries to self-contained LLVM IR. Our evaluation shows that Biotite successfully translates all benchmarks in SPEC CPU 2017 and achieves a 2.346× performance improvement over QEMU for the integer benchmark suite.
- 25 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 45 (2) , 90-100
Flexible Approximate Computing for Mitigating Branch Divergence in GPUs
Graphics Processing Units (GPUs) have become prevalent across various domains due to their high throughput and energy efficiency. A major challenge in GPU computing is branch divergence, which occurs when the control flow paths of concurrently executing threads diverge. Branch divergence significantly degrades performance, as diverged threads are executed in a serialized, time-sliced manner. To mitigate this issue, we propose a novel approximate computing technique for GPUs. Our method approximates the results of diverged threads by selectively terminating their execution. Furthermore, it incorporates a dynamic mechanism that adjusts both the probability of termination and the number of instructions executed before termination, based on user-defined parameters. This technique allows users to balance execution speed and result accuracy, offering greater flexibility in controlling computational quality. Additionally, by changing the approximation aggressiveness at runtime, we can accommodate applications such as VR/AR, where the demand for accuracy changes in real-time.
- 25 November 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Micro
- Vol. 45 (2) , 90-100
Flexible Approximate Computing for Mitigating Branch Divergence in GPUs
Graphics Processing Units (GPUs) have become prevalent across various domains due to their high throughput and energy efficiency. A major challenge in GPU computing is branch divergence, which occurs when the control flow paths of concurrently executing threads diverge. Branch divergence significantly degrades performance, as diverged threads are executed in a serialized, time-sliced manner. To mitigate this issue, we propose a novel approximate computing technique for GPUs. Our method approximates the results of diverged threads by selectively terminating their execution. Furthermore, it incorporates a dynamic mechanism that adjusts both the probability of termination and the number of instructions executed before termination, based on user-defined parameters. This technique allows users to balance execution speed and result accuracy, offering greater flexibility in controlling computational quality. Additionally, by changing the approximation aggressiveness at runtime, we can accommodate applications such as VR/AR, where the demand for accuracy changes in real-time.
- Open Access
- 13 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
DataPipettor: Touch-Based Information Transfer Interface Using Proximity Wireless Communication
As diverse computing devices, including wearable and embedded systems, become ubiquitous in our living spaces, sharing and transferring data between these computers has become a usual operation. This paper introduces DataPipettor, an information transfer interface that enables intuitive data movement based on touch interactions. DataPipettor utilizes proximity wireless communication (PWC), allowing high-speed real-time communication between proximate channels while simultaneously achieving touch sensing and data transmission/reception. The interface can be miniaturized due to the characteristics of PWC, making it suitable for use in wearable devices and small equipment interfaces. Users wearing the interface can intuitively transfer data through touch interactions as if physically passing objects. This paper conducted the development and evaluation of the interface, a conceptual demonstration, and discuss future prospects.
- 13 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-3
Open AccessDataPipettor: Touch-Based Information Transfer Interface Using Proximity Wireless Communication
As diverse computing devices, including wearable and embedded systems, become ubiquitous in our living spaces, sharing and transferring data between these computers has become a usual operation. This paper introduces DataPipettor, an information transfer interface that enables intuitive data movement based on touch interactions. DataPipettor utilizes proximity wireless communication (PWC), allowing high-speed real-time communication between proximate channels while simultaneously achieving touch sensing and data transmission/reception. The interface can be miniaturized due to the characteristics of PWC, making it suitable for use in wearable devices and small equipment interfaces. Users wearing the interface can intuitively transfer data through touch interactions as if physically passing objects. This paper conducted the development and evaluation of the interface, a conceptual demonstration, and discuss future prospects.
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 23 (2) , 211-214
Cycle-Oriented Dynamic Approximation: Architectural Framework to Meet Performance Requirements
Approximate computing achieves shorter execution times and reduced energy consumption in areas where precise computation written in a program is not essential to meet a goal. When applying the approximations, it is vital to satisfy the required quality-of-service (QoS) (execution time) and quality-of-results (QoR) (output accuracy). Existing methods have difficulty in maintaining a constant QoS or impose a burden on programmers. In this study, we propose the Cycleoriented Dynamic Approximation (CODAX) algorithms and processor architecture that minimize the burden on the programmer and maintain the execution time close to the required QoS while providing the user with an option to satisfy their QoR requirement. CODAX operates based on a threshold that indicates the maximum number of cycles available for one loop iteration. The longer the execution time of an iteration, the more aggressive the approximation. The threshold automatically increases or decreases at runtime to bring the total number of elapsed cycles close to the required QoS. Furthermore, CODAX allows the user to change the threshold to indirectly guarantee the required QoR. Our simulation revealed that CODAX brought the actual number of executed cycles close to the expected number for four workloads
- 6 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Computer Architecture Letters
- Vol. 23 (2) , 211-214
Cycle-Oriented Dynamic Approximation: Architectural Framework to Meet Performance Requirements
Approximate computing achieves shorter execution times and reduced energy consumption in areas where precise computation written in a program is not essential to meet a goal. When applying the approximations, it is vital to satisfy the required quality-of-service (QoS) (execution time) and quality-of-results (QoR) (output accuracy). Existing methods have difficulty in maintaining a constant QoS or impose a burden on programmers. In this study, we propose the Cycleoriented Dynamic Approximation (CODAX) algorithms and processor architecture that minimize the burden on the programmer and maintain the execution time close to the required QoS while providing the user with an option to satisfy their QoR requirement. CODAX operates based on a threshold that indicates the maximum number of cycles available for one loop iteration. The longer the execution time of an iteration, the more aggressive the approximation. The threshold automatically increases or decreases at runtime to bring the total number of elapsed cycles close to the required QoS. Furthermore, CODAX allows the user to change the threshold to indirectly guarantee the required QoR. Our simulation revealed that CODAX brought the actual number of executed cycles close to the expected number for four workloads
- Open Access
- 11 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-5
CommuTiles: Shape-Changeable Modular Computer System Using Proximity Wireless Communication
We present CommuTiles, a computer system where multiple tile-shaped modules are wirelessly connected. This represents the first implementation of the concept of a shape-changeable computer system capable of realizing diverse shapes and functions, as well as their dynamic changes. We designed tile-shaped modules equipped with wireless communication circuits and coils, and network protocols for data communication and coordinate estimation. By combining multiple tile-shaped modules with different functions, we implemented application examples. Through the implementation of hardware, software, and application cases, we lay the foundation for future research on shape-changeable computers and the shape-changing interfaces they enable.
- 11 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-5
Open AccessCommuTiles: Shape-Changeable Modular Computer System Using Proximity Wireless Communication
We present CommuTiles, a computer system where multiple tile-shaped modules are wirelessly connected. This represents the first implementation of the concept of a shape-changeable computer system capable of realizing diverse shapes and functions, as well as their dynamic changes. We designed tile-shaped modules equipped with wireless communication circuits and coils, and network protocols for data communication and coordinate estimation. By combining multiple tile-shaped modules with different functions, we implemented application examples. Through the implementation of hardware, software, and application cases, we lay the foundation for future research on shape-changeable computers and the shape-changing interfaces they enable.
- Open Access
- 7 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 316-317
Multi-Tree Network Protocol Enabling System Partitioning for Shape-Changeable Computer System
Shape-changeable computer system is proposed as a system that forms various shapes by communicating wirelessly with many adjacent chips. A network that supports diverse shapes and dynamic chip replacement is important, and methods for constructing ad-hoc wireless networks and enabling dynamic reconfiguration of the system at runtime have been proposed so far. However, there is room for improvement in performance because they are based on up*/down* routing. Moreover, system partitioning is required by applications such as micro-robots and shape-changing user interfaces, and no network that realizes this has been studied yet. In this study, we propose a multi-tree network construction method for improving routing performance and a protocol that enables system partitioning, and we verify them by simulation using a newly-developed simulator.
- 7 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 316-317
Open AccessMulti-Tree Network Protocol Enabling System Partitioning for Shape-Changeable Computer System
Shape-changeable computer system is proposed as a system that forms various shapes by communicating wirelessly with many adjacent chips. A network that supports diverse shapes and dynamic chip replacement is important, and methods for constructing ad-hoc wireless networks and enabling dynamic reconfiguration of the system at runtime have been proposed so far. However, there is room for improvement in performance because they are based on up*/down* routing. Moreover, system partitioning is required by applications such as micro-robots and shape-changing user interfaces, and no network that realizes this has been studied yet. In this study, we propose a multi-tree network construction method for improving routing performance and a protocol that enables system partitioning, and we verify them by simulation using a newly-developed simulator.
- Open Access
- 11 February 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Shape Estimation Algorithm for Collective Shape-Changing Interface Using Wirelessly Connected Computers
In recent years, research on shape-changing user interfaces composed of collective objects has been conducted extensively. With advancements in semiconductor technology, there is a prospect of miniature wireless-connected computers cooperating to achieve interaction through shape changes. To realize this prospect, the issue of shape estimation among multiple computers becomes crucial. Given the large number of computers involved, it’s more efficient to limit communication to nearby nodes. Ideally, shape estimation should be performed using only neighboring nodes. However, in such a model, where all distances are equal, solving the shape estimation problem from a network graph is known to be NP-hard. In this study, we propose a method that enables shape estimation by imposing constraints on the physical layout, using information from neighboring nodes only. Our approach achieves initialization in O(n) and addition or disconnection in O(1). Shape estimation is expected to facilitate more intuitive interaction with users and support the construction of high-throughput networks. To validate the proposed algorithm, we created a prototype of a shape-changing display and demonstrated shape estimation. This validation underscores the algorithm’s potential as a cornerstone for future shape-changing interface technologies.
- 11 February 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessShape Estimation Algorithm for Collective Shape-Changing Interface Using Wirelessly Connected Computers
In recent years, research on shape-changing user interfaces composed of collective objects has been conducted extensively. With advancements in semiconductor technology, there is a prospect of miniature wireless-connected computers cooperating to achieve interaction through shape changes. To realize this prospect, the issue of shape estimation among multiple computers becomes crucial. Given the large number of computers involved, it’s more efficient to limit communication to nearby nodes. Ideally, shape estimation should be performed using only neighboring nodes. However, in such a model, where all distances are equal, solving the shape estimation problem from a network graph is known to be NP-hard. In this study, we propose a method that enables shape estimation by imposing constraints on the physical layout, using information from neighboring nodes only. Our approach achieves initialization in O(n) and addition or disconnection in O(1). Shape estimation is expected to facilitate more intuitive interaction with users and support the construction of high-throughput networks. To validate the proposed algorithm, we created a prototype of a shape-changing display and demonstrated shape estimation. This validation underscores the algorithm’s potential as a cornerstone for future shape-changing interface technologies.
- Open Access
- 1 December 2023
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E106.D (12) , 1988-1996
FPGA-based Garbling Accelerator with Parallel Pipeline Processing
As more and more programs handle personal information, the demand for secure handling of data is increasing. The protocol that satisfies this demand is called Secure function evaluation (SFE) and has attracted much attention from a privacy protection perspective. In two-party SFE, two mutually untrustworthy parties compute an arbitrary function on their respective secret inputs without disclosing any information other than the output of the function. For example, it is possible to execute a program while protecting private information, such as genomic information. The garbled circuit (GC) — a method of program obfuscation in which the program is divided into gates and the output is calculated using a symmetric key cipher for each gate — is an efficient method for this purpose. However, GC is computationally expensive and has a significant overhead even with an accelerator. We focus on hardware acceleration because of the nature of GC, which is limited to certain types of calculations, such as encryption and XOR. In this paper, we propose an architecture that accelerates garbling by running multiple garbling engines simultaneously based on the latest FPGA-based GC accelerator. In this architecture, managers are introduced to perform multiple rows of pipeline processing simultaneously. We also propose an optimized implementation of RAM for this FPGA accelerator. As a result, it achieves an average performance improvement of 26% in garbling the same set of programs, compared to the state-of-the-art (SOTA) garbling accelerator.
- 1 December 2023
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E106.D (12) , 1988-1996
Open AccessFPGA-based Garbling Accelerator with Parallel Pipeline Processing
As more and more programs handle personal information, the demand for secure handling of data is increasing. The protocol that satisfies this demand is called Secure function evaluation (SFE) and has attracted much attention from a privacy protection perspective. In two-party SFE, two mutually untrustworthy parties compute an arbitrary function on their respective secret inputs without disclosing any information other than the output of the function. For example, it is possible to execute a program while protecting private information, such as genomic information. The garbled circuit (GC) — a method of program obfuscation in which the program is divided into gates and the output is calculated using a symmetric key cipher for each gate — is an efficient method for this purpose. However, GC is computationally expensive and has a significant overhead even with an accelerator. We focus on hardware acceleration because of the nature of GC, which is limited to certain types of calculations, such as encryption and XOR. In this paper, we propose an architecture that accelerates garbling by running multiple garbling engines simultaneously based on the latest FPGA-based GC accelerator. In this architecture, managers are introduced to perform multiple rows of pipeline processing simultaneously. We also propose an optimized implementation of RAM for this FPGA accelerator. As a result, it achieves an average performance improvement of 26% in garbling the same set of programs, compared to the state-of-the-art (SOTA) garbling accelerator.
- Open Access
- 1 December 2023
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E106.D (12) , 1960-1968
A Principal Factor of Performance in Decoupled Front-End
One of the performance bottlenecks of a processor is the front-end that supplies instructions. Various techniques, such as cache replacement algorithms and hardware prefetching, have been investigated to facilitate smooth instruction supply at the front-end and to improve processor performance. In these approaches, one of the most important factors has been the reduction in the number of instruction cache misses. By using the number of instruction cache misses or derived factors, previous studies have explained the performance improvements achieved by their proposed methods. However, we found that the number of instruction cache misses does not always explain performance changes well in modern processors. This is because the front-end in modern processors handles subsequent instruction cache misses in overlap with earlier ones. Based on this observation, we propose a novel factor: the number of miss regions. We define a region as a sequence of instructions from one branch misprediction to the next, while we define a miss region as a region that contains one or more instruction cache misses. At the boundary of each region, the pipeline is flushed owing to a branch misprediction. Thus, cache misses after this boundary are not handled in overlap with cache misses before the boundary. As a result, the number of miss regions is equal to the number of cache misses that are processed without overlap. In this paper, we demonstrate that the number of miss regions can well explain the variation in performance through mathematical models and simulation results. The results show that the model explains cycles per instruction with an average error of 1.0% and maximum error of 4.1% when applying an existing prefetcher to the instruction cache. The idea of miss regions highlights that instruction cache misses and branch mispredictions interact with each other in processors with a decoupled front-end. We hope that considering this interaction will motivate the development of fast performance estimation methods and new microarchitectural methods.
- 1 December 2023
- Published by Institute of Electronics, Information and Communications Engineers (IEICE) in IEICE Transactions on Information and Systems
- Vol. E106.D (12) , 1960-1968
Open AccessA Principal Factor of Performance in Decoupled Front-End
One of the performance bottlenecks of a processor is the front-end that supplies instructions. Various techniques, such as cache replacement algorithms and hardware prefetching, have been investigated to facilitate smooth instruction supply at the front-end and to improve processor performance. In these approaches, one of the most important factors has been the reduction in the number of instruction cache misses. By using the number of instruction cache misses or derived factors, previous studies have explained the performance improvements achieved by their proposed methods. However, we found that the number of instruction cache misses does not always explain performance changes well in modern processors. This is because the front-end in modern processors handles subsequent instruction cache misses in overlap with earlier ones. Based on this observation, we propose a novel factor: the number of miss regions. We define a region as a sequence of instructions from one branch misprediction to the next, while we define a miss region as a region that contains one or more instruction cache misses. At the boundary of each region, the pipeline is flushed owing to a branch misprediction. Thus, cache misses after this boundary are not handled in overlap with cache misses before the boundary. As a result, the number of miss regions is equal to the number of cache misses that are processed without overlap. In this paper, we demonstrate that the number of miss regions can well explain the variation in performance through mathematical models and simulation results. The results show that the model explains cycles per instruction with an average error of 1.0% and maximum error of 4.1% when applying an existing prefetcher to the instruction cache. The idea of miss regions highlights that instruction cache misses and branch mispredictions interact with each other in processors with a decoupled front-end. We hope that considering this interaction will motivate the development of fast performance estimation methods and new microarchitectural methods.