IA et big data : CASD-IA

CASD-IA

CASD can make available on request environments, certified ISO 27 001, HDS and RGPD, designed specifically for projects requiring IA infrastructure and technologies: CASD-IA.

CASD-IA can provide hardware and software tools to meet, on request and after study, the needs of projects involving sensitive data, whether in terms of security, massive data storage or computing power. The entire chain of installation and start-up of clusters as well as the maintenance of their software resources is carried out by the CASD-IA team. CASD’s “AI and big data” on-demand services aim to accelerate research, innovation and production in the field of Artificial Intelligence and massive data.

The CASD-IA compartment distinguishes itself from existing solutions by placing particular emphasis on security management: thanks to a patented technological isolation to secure data by access via a dedicated box: the SD-Box ; the management of the traceability of treatments ; the integration of data processing technologies: mllib, machine learning, deep learning…

The provision of a very high and homogeneous level of security allows very high potential data crossover possibilities.

The CASD-IA compartment is a service open to innovation in terms of datascience.

Use CASD-IA :

CASD is at your disposal : if you wish to experiment these new technologies on your data ; if you want to work on a business partner’s data and want to provide guarantees regarding data security. We can provide answers in this case as a trusted third party.
Do not hesitate to contact us service@casd.eu

CASD-IA offer

CASD-IA offers, on demand after a study of the need, a platform of secure services, certified ISO 27 001, HDS and RGPD, and customized for calculations and processing in the field of AI and big data.

The platform operated by the project may have tools and services for data analysis or visualization. For example :

statistics (Ex: R-Evolution),
machine learning (MLlib, MAHOUT, SPARK)
deep Learning (TensorFlow…)
generalists (e.g. Python),
mining (e.g. Knime),
graphics (e.g. Dataviz, R-shiny, Bokeh)
in general access to Appstore analytics
Prior to implementation, the CASD establishes service specifications based on the expressed needs of the project.

Use CASD-IA :

The CASD is at your disposal:

if you wish to experiment with these new technologies on your data
if you want to work on a business partner’s data and want to provide guarantees regarding data security. We can provide answers in this case as a trusted third party.
Do not hesitate to contact us service@casd.eu

Infrastructure IA : CASD-IA

IA infrastructure to meet the most demanding needs.

Hardware configurations can include physical computers, storage devices and networks whose architecture is adapted to the computing and storage power requirements for AI and big data.
These configurations, physically installed in France, provide all the guarantees of confidentiality required by data owners and control authorities, particularly in terms of ISO 27 001, HDS and RGPD certification.

An ultra-secure CASD-IA compartment (CASD technology) can be set up on request after a study of the needs. It will be housed in a bubble isolated from the external network, accessible only via the SD-Box of the CASD. The servers within this compartment are connected to each other via a network link sufficiently sized for storage and distributed processing needs. The hardware configuration in terms of vCPU, RAM and storage is adaptable.

The software configuration is adaptable to project requirements.

The resources generally implemented by the projects are based on Hadoop (including Cloudera and HortonWorks distributions), Hive, Impala, Sqoop, MongoDB, Neo4j, ElasticSearch, Parstream, or Redis.

As far as AI tools are concerned, there are adaptable infrastructures for each project with tools such as Python, TensorFlow, MXNet, Theano, Keras, Caffe, ConvNet, PyTorch, Anaconda …

The deployment of this software infrastructure is carried out after study with the applicant and can be incremental to meet the expectations of the projects and the opportunities offered by new technologies appearing at a very fast pace.