Guide System Services

04

01

03

02

Spark / Hadoop as a parallel computing environment

Loosely coupled integrated architecture of HPDA/AI environment with HPC environment

Architecture for AI processing and analysis of primary data by HPDA system with Big Data Discovery

Architecture for data collection, transaction processing, AI processing and IoT analysis in HPDA/AI

10

06

07

05

Monte Carlo modeling

Fast Fourier Transformation algorithm

Markov chains modeling

Application Services

Service 2

Home

Main purpose

Loosely coupled integrated architecture of HPDA/AI environment with HPC environment

• To present possible integration between HPC environment and HPDA environment, via different networking features and types.

The HPC/HPDA/AI environment can be separated in 2 environments – HPC and HPDA/AI environment. The HPC environment in the extreme way is a Supercomputer system. The HPDA/AI environment in the normal way is a Hadoop system.

+ info

Service 3

Service 1

Main purpose

Architecture for data collection, transaction processing, AI processing and IoT analysis in HPDA/AI

• To present a software architecture with capabilities of collection, transacting and processing data, implementing AI as well.

Generally, a system Hadoop with its ecosystem consists of many processors, capable of parallel high performance processing, containing many tools for Data analysis and tools for AI. For this reason, the Hadoop system can be considered as HPDA/AI environment.

+ info

Service 4

Service 2

Main purpose

• To present a functional architecture of the Big Data Discovery features.

Three levels of Big Data Discovery:

• Level 1: Working mainly on structured and semi-structured data
• Level 2: Working on non-structured data with using of AI
• Level 3: Developing AI based analysis from scratch, using different engines for AI

Architecture for AI processing and analysis of primary data by HPDA system with Big Data Discovery

+ info

+ Level 3

+ Level 1

+ Level 2

Service 5

Service 3

Main purpose

• To present a software/hardware architecture ensuring mass parallel programs running in HPDA environment with the help of Spark.

Spark/Hadoop is a platform realized with MapReduce for large-scale data processing. The execution time of Spark application is a significant factor in measuring real-time processing especially with large big data files.

Spark / Hadoop as a parallel computing environment

+ info

Service 6

Service 4

Main purpose

• To present a software component supporting parallel execution of Markov chains models in HPDA environment.

The Markov chains modeling can be used as a component of HPDA analytical toll. A Markov chain or Markov process is a stochastic model describing a sequence of possible events in which the probability of each event depends only on the state attained in the previous event.

Markov chains modeling

+ info

Service 7

Service 5

Main purpose

• To present a software component supporting parallel execution of Monte Carlo models in HPDA environment.

The Monte Carlo modeling can be used as a component of HPDA analytical toll. Monte Carlo is one of many simulation/modeling types that execute a huge amount of repetitive tasks that use relatively little data.

Monte Carlo modeling

+ info

HOME

Service 6

Main purpose

• To present a software component supporting parallel execution of Fast Fourier Transformation in HPDA environment.

The Fast Fourie Transformation algorithm can be used as a component of HPDA analytical toll. Fast Fourier transform has an extremely important position in the field of scientific computing.

Fast Fourier Transformation algorithm

+ info