WASAI  Big  Data

Processing  Accelerator

WASAI Big Data acceleration platform leverages the power of FPGAs. Traditional CPUs are usually built with a relatively small number of processing cores under von Neumann architecture. FPGA's reconfigurable nature allows us to design parallel architectures consisting of thousands of smaller and high-efficiency cores optimized for specialized computing purpose.

WASAI Big Data acceleration platform provides this adaptable, flexible and programmable parallelism to existing software ecosystems transparently, such as Apache Hadoop MapReduce, Apache Hive and Apache Spark. We provide compatible software programming interface which is familiar to all existing software programmers and no HDL nor OpenCL programming is required.

Performance is a key concern for all software systems. WASAI's most efficient and optimized solutions are faster, longer-lasting and have lower power and thermal requirements than those available on the market. One single node of WASAI Big Data accelerator is faster than a cluster of CPU servers and a rack of WASAI Big Data accelerators is faster than a data center solution.


Contact us now to ensure your Big Data analytics performs to its full potential!

Product  Overview

WASAI Mapreduce Accelerator

WASAI MapReduce Accelerator is built on FPGAs with innovative parallelism up to thousands of processing cores and optimized pipelining.

WASAI Hive/Spark SQL Accelerator

WASAI Hive & SparkSQL Accelerator leverages FPGAs and a state-of-the-art query execution planner to unleash the system's potential.

WASAI BIG DATA Solution Real Benchmark

Big Data acceleration benchmarks

Traditional software solutions are usually built with a relatively small number of CPU cores under von Neumann architecture, obviously consuming a great deal of energy.

Within the same period of time and platform environment, WASAI Big Data Accelerator with FPGA only consumes 1/3 of energy and CPU resources.