Autor: SAAL Operating Systems Sole Proprietorship LLC
Integrated Big Data & AI Platform
Integrated Big Data & AI Platform with 100+ Use Cases & Arabic AI tools.
DigiXT is a low code/ no code integrated big data & AI platform, DigiXT architecture is modular and supports different integrations.
Among DigiXT’s key benefits is the design DNA that reduces dependency on skilled resources for data management & operation supporting the whole process of data ingestion, transformation, storage, and consumption. We have augmented DigiXT with ready-to-deploy AI tools to make it easy for users to build customized products, and are available in DigiXT inbuilt marketplace as add-ons for the users to buy and use, this will facilitates AI & big data use-cases development and implementation through an agile, easy and flexible approach.
Needless to mention the Arabic AI Capabilities DigiXT’s unique value proposition. We have invested in Arabic AI and developed a series of tools that can be used to enrich use-cases that require usage of AI to understand or process Arabic text, voice, and other format of data. Among the tools we count; Speech to Text and Text to Speech capabilities, Natural Language Processing, Arabic Sentiment Analysis, Pronunciation Verification, Anomaly Detection, Automatic Diacritization, Spelling Validation, Arabic Question Answering, Arabic Topic classification, Arabic Text Summarization Etc ..
DigiXT Architecture : we call it 5 Cs architecture; Connect, Construct, Contain, Consume, and Control.
Connect layer : provides more than 200 connectors for data ingestion and supports many type of ingestion (stream, batch, API) as well as many type of data (structured, unstructured, semi-structured, real time data …), it also supports CDC where it stream changes without directly querying the underlying source..
Construct layer : As opposed to loading only relevant data (ETL), DigiXT uses ELT – a more comprehensive way of loading “ALL” data. This layer is powered by more than 600 extraction functions allowing the users to apply the most common “Data Wrangling” and “Data Munching” to clean and transform “ALL” data.
CONTAIN Layer : This storage layer of DigiXT uses the most efficient storage powered by Arrow and ORC (Arrow is a cross-platform data format and ORC stands for Optimized Row Columnar format for fast retrieval). In regards to the type of storage, DigiXT support all type of data storage including object storage and allows consumption from external storage as well.
CONSUME Layer : The Consume layer's role is to be a secured centralized point of data access by different parts of the platform. Consuming apps, Machine Learning solutions, APIs, and external tools are examples of possible data consumers might need to retrieve the information. It also brings the capability of building dashboards and creating powerful visualizations of the data.
CONTROL Layer : The control layer is the main entry point for managing the resources of the platform. From the dashboard, admin users can follow the cluster's health, launch and monitor components, get diagnostics, audit changes, as well as manage users and accesses.
All five layers are architected as “PLUGINS”, which allows the replacement of any layer when a new trend or technology needs to be incorporated. The layers are packaged within the strong realm of security, audit, management, and monitoring so that the OPS team has full transparency to “WHO” and “WHAT”. All communications including data in motion and data at transit are encrypted and through HTTPS. Custom authentication mechanisms allow fine-grained access.
Lastly, DigiXT supports all hosting models as well as hybrid models.