Bodo is the next generation large scale data and AI compute platform that provides major improvement in performance, scalability, simplicity and cost over other solutions. Bodo’s technology is a culmination of years of research by high performance computing (HPC) and compiler experts, achieving what was thought of as impossible.
Data processing frameworks have historically presented a trade-off between simplicity of high-level languages like Python and performance of HPC paradigms like MPI/C++. Bodo achieves both simultaneously using the first automatic parallelization and optimization compiler technology.
Bodo engine incorporates the first inferential compiler – it can infer the structure of the application to optimize and parallelize it automatically. This is as if an HPC expert rewrote the code in a low-level paradigm, but it happens transparently and in real time.
Bodo is the first engine that provides the full parallelism and theoretically unlimited scalability of Single Program Multiple Data (SPMD) architecture. Previous engines approximate parallelism using master-executor libraries, which incur significant task overheads and sequential bottlenecks.
Check out our Blog.