IoT/M2M log data processing platform created with GPU. Optimizing the data lifecycle.
"PG-Strom" is an open-source extension module that fully leverages the capabilities of GPUs and NVMe to accelerate PostgreSQL at lightning speed with the power of thousands of cores operating in parallel, tackling massive amounts of data exceeding terabytes. GPUs are equipped with thousands of processor cores designed to execute parallel computations extremely efficiently. Additionally, it allows for direct execution of statistical analysis algorithms on transaction data stored in the database. These logics can be computed rapidly by the GPU, eliminating the need to export data with each calculation. 【Features】 ■ Architecture that maximizes the performance of high-speed storage ■ Eliminates the need for data import before aggregation and search ■ Ensures that experience and skills with PostgreSQL are not wasted *For more details, please download the PDF or feel free to contact us.
Inquire About This Product
basic information
【Other Features】 ■ Search and aggregate the ever-increasing IoT/M2M log data with overwhelming processing efficiency ■ Real-time location analysis made possible by GPU ■ Execute packet capture and search in an all-in-one solution ■ Business intelligence & reporting ■ Integration with machine learning / anomaly detection, etc. *For more details, please download the PDF or feel free to contact us.
Price range
Delivery Time
Applications/Examples of results
【For concerns like these】 ■ It takes a long time to import log data before analysis ■ The aggregation process is unbearably slow ■ Will employees be able to effectively learn to use the new tools? *For more details, please download the PDF or feel free to contact us.
catalog（1）
Download All Catalogs
Company information
Our company develops and maintains software products for large-scale data processing. Additionally, we offer professional services such as database tuning services, R&D support services, and GPU software development. In 2012, when the idea of applying GPUs to database processing was still almost nonexistent and no one knew how to design the software, we released a prototype that would become the foundation of the current PG-Strom, and we have since gone through various trials and errors. Please feel free to contact us if you have any inquiries.