International audienceCustom hardware accelerators usage is shifting towards new application domains such as graph analytics and unstructured text analysis. These applications expose complex control-flow which is challenging to map to hardware, especially when operating from a C/C++ description using High-Level Synthesis toolchains. Several approaches relying on speculative execution have been proposed to overcome those limitations, but they often fail to handle the multiple interacting speculations required for realistic use-cases. This paper proposes a fully automated hardware synthesis flow based on a source-to-source compiler that identifies and explores intricate speculation configurations to generate speculative hardware accelerators
The major specific contributions are: (1) We introduce a new compiler analysis to identify the memor...
Producción CientíficaSoftware-based, thread-level speculation (TLS) is a software technique that opt...
International audienceNowadays almost every device has parallel architecture, hence parallelization ...
International audienceCustom hardware accelerators usage is shifting towards new application domains...
International audienceLoop pipelining is a key optimization in modern HLS tools for synthesizing eff...
International audienceThe RISC-V ecosystem is quickly growing and has gained a lot of traction in th...
High Level Synthesis (HLS) is a process which, starting from a high-level description of an applicat...
Specialized accelerators can exploit spatial parallelism on both operations and data thanks to a ded...
International audienceFPGAs are well known for their ability to perform non-standard computations no...
Effectively utilizing available parallelism is becoming harder and harder as systems evolve to many-...
International audienceDesigning FPGA-based accelerators is a difficult and time-consuming task which...
The demand for scalable, high-performance computing has increased as the size of datasets has grown ...
Machine learning algorithms continue to receive significant attention from industry and research. As...
High-level synthesis (HLS) is a key component for the hardware acceleration of applications, especi...
International audienceHigh-level synthesis (HLS) is a big step forward in terms of design productivi...
The major specific contributions are: (1) We introduce a new compiler analysis to identify the memor...
Producción CientíficaSoftware-based, thread-level speculation (TLS) is a software technique that opt...
International audienceNowadays almost every device has parallel architecture, hence parallelization ...
International audienceCustom hardware accelerators usage is shifting towards new application domains...
International audienceLoop pipelining is a key optimization in modern HLS tools for synthesizing eff...
International audienceThe RISC-V ecosystem is quickly growing and has gained a lot of traction in th...
High Level Synthesis (HLS) is a process which, starting from a high-level description of an applicat...
Specialized accelerators can exploit spatial parallelism on both operations and data thanks to a ded...
International audienceFPGAs are well known for their ability to perform non-standard computations no...
Effectively utilizing available parallelism is becoming harder and harder as systems evolve to many-...
International audienceDesigning FPGA-based accelerators is a difficult and time-consuming task which...
The demand for scalable, high-performance computing has increased as the size of datasets has grown ...
Machine learning algorithms continue to receive significant attention from industry and research. As...
High-level synthesis (HLS) is a key component for the hardware acceleration of applications, especi...
International audienceHigh-level synthesis (HLS) is a big step forward in terms of design productivi...
The major specific contributions are: (1) We introduce a new compiler analysis to identify the memor...
Producción CientíficaSoftware-based, thread-level speculation (TLS) is a software technique that opt...
International audienceNowadays almost every device has parallel architecture, hence parallelization ...