
Klepsydra AI Inference Add-On
Klepsydra Technologies
The Klepsydra High-Performance Software Framework consists of the following elements:
Klepsydra Streaming
Boost data processing for on-board processors for general applications and processor intensive algorithms, resulting in low latency, 10x more data processing capabilities and 50% power reduction
GPU / FPGA Connector
High parallelisation on GPU to increase the processing data rate and GPU utilisation
Easy integration of FPGA HW acceleration
ROS2 Executor plugin
Executor for ROS2 able to process up to 10 x more data with up to 50% reduction in CPU consumption.
Klepsydra Streaming Distribution Optimiser (SDO)
Webtool to optimise algorithm pipelines for maximum data throughput. Klepsydra SDO can optimise algorithms running on CPU, GPU, or FPGA
Parameter | Value |
---|---|
Lead Time | 2 to 3 weeks |
Operating System | Linux |
Operating System | FreeRTOS |
Operating System | RTEMS |
Supported Processor | ARM Cortex A7 |
Supported Processor | ARM Cortex A9 |
Supported Processor | ARMv8 |
Supported Processor | Cobham Leon3 |
Supported Processor | x86, 64 bit or 32 bit |
Supported Languages | C++ |
Supported Languages | C |
Supported Languages | NodeJS |
Supported Languages | Python |
Supported Third Party Plugins | ROS, ROS2, DDS, ZMQ, CAN |
Frameworks | MATLAB, PX4, MAVROS |
The Klepsydra High-Performance Software Framework consists of the following elements:
Klepsydra Streaming
Boost data processing for on-board processors for general applications and processor intensive algorithms, resulting in low latency, 10x more data processing capabilities and 50% power reduction
GPU / FPGA Connector
High parallelisation on GPU to increase the processing data rate and GPU utilisation
Easy integration of FPGA HW acceleration
ROS2 Executor plugin
Executor for ROS2 able to process up to 10 x more data with up to 50% reduction in CPU consumption.
Klepsydra Streaming Distribution Optimiser (SDO)
Webtool to optimise algorithm pipelines for maximum data throughput. Klepsydra SDO can optimise algorithms running on CPU, GPU, or FPGA