FAQ Klepsydra FAQ

FAQ

What is Klepsydra?

Klepsydra is a leading software company that specializes in developing cutting-edge data processing solutions for real-time applications. By utilizing state-of-the-art technology, Klepsydra is able to offer processing speeds that are up to 8 times faster than traditional methods, all while consuming 50% less energy. This makes Klepsydra’s solutions both highly efficient and cost-effective, making it the ideal choice for companies that demand high-performance data processing capabilities.

Does Klepsydra distribute/manufacture hardware?

No, the company does not manufacture or distribute hardware, its cutting-edge software has been designed to enable heavy data processing and analytics at lightning-fast speeds.

Which kind of data can Klepsydra process?

Klepsydra’s data processing solutions are agnostic to the type of data being processed, meaning they can handle data from any kind of sensor. Whether it’s Images, LiDAR data, SAR data or time series,
Klepsydra’s software is designed to efficiently process and analyse large amounts of data. In fact, the more data that is processed, the better the performance of Klepsydra’s solutions.

How does Klepsydra work?

Klepsydra is a highly efficient software algorithm that utilises parallel processing techniques inspired by the financial industry to process data.

How does Klepsydra compare to ROS 2?

Klepsydra is a compatible tool for accelerating data processing that integrates seamlessly with the ROS 2 Executor through a plug-in. With this plug-in, Klepsydra is able to run up to 10 times faster while consuming 50-70% less power than ROS2 alone. While Klepsydra is not a direct competitor to ROS2, it offers a powerful and complementary solution for those looking to accelerate their data processing capabilities.

In which language does Klepsydra work?

Klepsydra’s main API is written in C++11. However, Klepsydra also offers bindings for Node JS, C, and Matlab, making it accessible to a wider range of developers and applications. Klepsydra makes it easier to integrate its solutions into their existing tech stacks and workflows, helping them to quickly and efficiently take advantage of Klepsydra’s powerful data processing capabilities.

Which are the requirements for Klepsydra?

Klepsydra requires three key components in order to run in a computer:

  1. Operating System: Klepsydra is compatible with a variety of operating systems, including Linux, Real-Time Multi-Tasking Operating System (RTOS), and FreeRTOS. The specific choice of operating system will depend on the requirements of the target system and the preferences of the developer.
  2. Atomic Operations: Klepsydra relies on hardware support for atomic operations, which are executed as a single, indivisible unit of work. This is essential for ensuring that concurrent threads can safely access shared resources without causing data inconsistencies or race conditions.
  3. C++11 Compiler: Klepsydra is written in C++, and it requires a C++11-compliant compiler for building and execution. This version of C++ introduces several new features and improvements to the language, such as improved type inference, lambda functions, and move semantics, which are leveraged by Klepsydra to provide efficient and scalable real-time messaging services.

In which hardware can Klepsydra be deployed?

Klepsydra is currently compatible with CPU, ARM, and x86 architectures. However, we are actively working on expanding its compatibility to include GPUs and FPGAs. GPU support for non-AI applications is available since Dec 2022, AI applications support is planned for Q2 2023. FPGA support is planned for Q3 2023.

Can I run AI on-board with Klepsydra?

Yes, Klepsydra has recently launched Klepsydra AI, an advanced inference engine solution that can run a wide variety of ML/AI algorithms on edge devices. Compared to leading competitors in the market, Klepsydra AI is capable of processing 2-6 times more data while utilizing up to 50% less processor consumption.

Can Klepsydra provide ML/AI algorithms?

Yes, Klepsydra provides support for training and developing customized ML algorithms related to object detection, vision navigation, and computer vision. With Klepsydra, you can access ad hoc ML algorithms tailored to meet your specific needs.

Where does Klepsydra sit on the product stack?

Klepsydra is a library that sits on top of the operating system, similar to other libraries. As such, it can be easily incorporated into your own application, allowing you to utilise its functionality seamlessly.

How is Klepsydra beneficial for the environment?

Klepsydra’s low power consumption results in computers running Klepsydra using less than half of their processing energy, which greatly reduces their environmental impact. Furthermore, Klepsydra’s ability to extend the lifespan of computers reduces the need for manufacturing new ones, which contributes to a more sustainable future.

How difficult is the installation of Klepsydra?

Installing Klepsydra is quick and easy. Simply follow the instructions provided in our tutorials, and in just 20 minutes, Klepsydra will be fully installed and ready to use on your computer.

Is Klepsydra a product for today’s technology?

Absolutely. With the rise of edge computing, Klepsydra is more essential than ever before. The amount of data generated by edge computing systems is massive, thanks to the use of advanced sensors and high-speed communication technologies such as 5G. To process this data in real-time, a tool like Klepsydra is necessary