AIHPC® Cloud - Empower Big Datasets with AI (Artificial Intelligence) HPC (High-Performance Computing)

Generic placeholder image

Expand HPC by Big Data Container Platforms

Designed and engineered for NASA/NOAA NextGen specifications on satellite image machine learning, AIHPC helps you expand the traditional HPC market by processing both HPC and Big Data workload with the enhanced AIHPC Spark and Hadoop container platforms and HPC cloud clusters

View details »

Generic placeholder image

Modernize HPC by AI Cloud Platforms

AIHPC helps you modernize HPC by processing imagery and computer vision workload with the enhanced AIHPC artificial intelligence platforms (ImageJ and OpenCV3) on HPC cloud infrastructure

View details »

Generic placeholder image

Container Parallel Processing and AI Defense

Use container parallel processing to make your dockerized cloud HPC node 5 times faster than a regular EC2 host; building on the ZDAF AMI with dual-layer security for DOS and ZeroDay AI defense

View details »


Next Steps: How to Set Up Your AIHPC® Cluster in 3 Easy Steps

  1. Launch the AIHPC cloud from AWS Cloud DeepCyber Catalog to an EC2 instance and obtain {Your New IP/DNS}
  2. SSH into the new EC2 and verify that "/home/ubuntu/lb.sh" launches 5 docker containers and a container load balancer
  3. Read AI 303 Training Manual; AIHPC works as a single EC2 host for non-HPC workload; you may also create your HPC cloud cluster with the AIHPC AMI as the custom AMI.

FAQs for AIHPC® (click questions for answers)

AIHPC AWS Product Page -> Continue -> Manual Launch -> AMI IDs for different regions

/home/ubuntu/lb.sh {execute this script from command line}

With AIHPC, the additional 5 Docker containers are running parallelly on the worker nodes. This makes your HPC 5 times faster than the one without using AIHPC for the worker nodes.

Yes. With AIHPC, each of the 5 containers of a worker node can process similar workload as an original worker node.

browser -> http://{Your IP/DNS}:9000/

docker ps {this verifies that the load balancer are running for the 5 containers}

See AIHPC extended FAQs. If you have procured the annual subscription of one AIHPC EC2, send a request (for the extended FAQs) with your AWS account number to demo@deepcybe.com.

1) cd ~/opencv-3.0.0/obama_face; 2) python t1.py haarcascade_frontalface_default.xml obama-phone.jpg; 3) cd /tmp; w3m obama-phone.jpg.faces.jpg

You need to use a X11 session to view the graphic result. See AIHPC extended FAQs for using the X11 session. If you have procured the annual subscription of one AIHPC EC2, send a request (for the extended FAQs) with your AWS account number to demo@deepcybe.com.

AIHPC is licensed on the numbers of EC2 instances. For a single derived server, you may keep a small original AIHPC EC2 up and running to pay for your derived AMI instance. For multiple derived AIHPC servers, you may procure an enterprise license and/or a support plan from demo@deepcybe.com.

Must use ubuntu, not root or ec2-user. For example, assuming add.pem is your key pair, the correct command is: ssh -i add.pem ubuntu@Your_AIHPC_IP

Satellite image machine learning: at present military hires many human eyes to track an object (e.g., a warship) in satellite images. In the future, AIHPC may train cloud HPC clusters with AI deep learning to identify and track objects in massive satellite images.

DeepCyber's Products

DC Aloe Series

AIHPC: next-generation AWS cloud product to help uncover the hidden values of your big datasets with Big Data container platforms, HPC cloud clusters, AI platforms, AI cyber defense and cloud container computing.

DSS: a group of enterprise Data Science servers in the cloud to enable predictive analytics, anomaly detection, factor and conjoint analytics for data-driven decisions.

DevOps: enables a decent CICD operation with a few clicks.

CES: Continuous Enterprise Security.

DC Celtis Series

AI Cyber: ZDAF uses AI cyber defense technology to effectively attenuate ZeroDay and DOS cyberattacks.

DC Pine Series

AI Investing: PNN automatically trains and validates deep learning models to predict trends of security prices

DC Howea Series

AI Health: PED Number app quantifies and boosts the level of your positive energy on a daily basis and predicts health to reduce insurance cost.


Email: demo@deepcybe.com.