Intel re-released Megh’s foundational paper on Megh Open Analytics Platform— “Real-time streaming analytics at the edge”—with performance validation results, concluding that the platform “redefines video analytics for outstanding performance.”
After extensive testing, Intel reported that validation of performance results of Megh Video Analytics Solution (VAS) running on Intel® NUC kits with 11th or 12th Gen Intel® Core™ processors demonstrated “impressive capabilities” to scale with sustained performance.
Megh developed Megh Open Analytics Platform to address the complex challenges in implementing edge analytics. It enables easy implementation of analytics pipelines and supports a model server to accelerate AI algorithms using CPUs, GPUs, FPGAs, and SOCs. Based on Open Analytics, an innovative approach to analytics emphasizing openness, the platform supports:
- Edge-to-cloud deployments across hardware architectures with a single, unified deployment framework.
- Continuous training of pre-trained AI models with new data.
- Contextual analytics, in which existing cameras and sensors are used to consider the context of the data.
- Cross-platform support, which abstracts underlying hardware and frameworks, allowing applications to be transportable and pipelines to be distributed across hardware vendors.
See the white paper to learn more about how Megh redefines real-time analytics to provide true operational reliability using AI+ on CPUs, GPUs, FPGAs, and SOCs from edge to cloud.
Also see our post, Megh VAS performance and validation report on Intel NUC kit.