Space Responsive Multithreaded Processor (SRMTP) for Spacecraft Control

Nakabeppu, Shota; Ide, Yosuke; Takahashi, Masahīko; Tsukahara, Yuta; Suzuki, Hiromi; Haruki, Shishido; Yamasaki, Nobuyuki

doi:10.1109/coolchips49199.2020.9097637

Cited by 9 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We apply our proposed hardware unit, the Data Rearrange Unit (DRU) covered in Chapter 4 to the Space Responsive Multithreaded Processor (SRMTP) SoC [15]. The overview of the SRMTP is shown in Fig.…”

Section: Targeted Processormentioning

confidence: 99%

“…The SRMTP is an embedded SoC for distributed real-time systems mainly focused on spacecraft control. It implements a Responsive Multithreaded Processing Unit (RMTPU) [15][19], which is an 8-way prioritized SMT processor. Spacecraft control requires real-time execution in which tasks have time constraints and need to be executed in priority order.…”

Section: Targeted Processormentioning

confidence: 99%

“…We also evaluate the impact of each of the three DRU interfaces and buffer size, which holds rearranged data. The DRU is implemented on the SRMTP [15], an 8-way prioritized Simultaneous Multi-Threading (SMT) processor for embedded real-time systems. We use the vector unit implemented on the SRMTP as the target data-parallel execution unit.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Data Rearrange Unit for Efficient Data Computation

Mamiya

Yamasaki

2022

IJNC

Self Cite

View full text Add to dashboard Cite

Recently, the demand for computation-intensive applications such as multimedia and AI applications has increased. Data-parallel execution units are typically used for calculations in these applications to increase computational throughput. However, the data required for computation may need to be accessed at discontinuous memory addresses, which can reduce computation efficiency.Generally, normal memory access instructions access data blocks of continuously allocated memory addresses, containing both valid and invalid data for computation. These memory access patterns result in low computation density in the data parallel execution units, wasting computation resources. Therefore, simply increasing the number of data-parallel execution units leads to an increase in wasted computational resources, which will become a significant issue in embedded systems where multiple resource limitations exist. It is essential to improve computational efficiency to perform practical computation in such systems.This paper introduces a Data Rearrange Unit (DRU), which gathers and rearranges valid computation data between main memory and execution units. The DRU improves the performance of multimedia and AI application by significantly reducing the access rate from/to main memory and increasing computation efficiency. It is applicable to most hardware architectures, and its effectiveness can be further enhanced by the execution unit interface that directly connects the DRU to the execution unit. We demonstrate the effectiveness of our DRU by implementation on the RMTP SoC, improving convolution throughput on a data-parallel execution unit by a maximum of 94 times while only increasing the total cell area by about 12.7%.

show abstract

Section: Targeted Processormentioning

confidence: 99%

Section: Targeted Processormentioning

confidence: 99%