Strato
Strato is an on-premise cloud platform based on servers with CPU and GPU components that enable you to solve both generic and specific computational problems using virtual machines that you can configure yourself. This can be for both short-term problem solving, as well as work that takes a long time.
What is Strato?
Strato is a Linux-based virtual computer platform to support data driven research that needs additional processing capacity. Network storage allows users to store and process large datasets, and the OpenStack platform allows the user to change the size of the computer (processing capacity) that is connected to their boot disk.
The Strato platform is a cluster of hardware that is virtualised to create instances that essentially function as a regular computer environment. Strato Instances are virtual machines, that can be launched by the user when they need it. When the instance has been created, it can be accessed from a terminal application on the user's local computer. The required software can then be installed and configured to the user's needs. By using Strato your research can benefit from additional RAM, CPU and GPU hardware. Strato can thus provide you with flexibility and additional computational power, when your research needs it.
Introduction to Strato
Basic information and instructions for first-time users of Strato.
- Training of models on GPUs
- Collaboration on code development
- Creation of Jupyter Notebook and RStudio Server hosts
- Simulation of physics and biomolecular models
- Testing of web applications
- Teaching of advanced computational methods
- Testing of distributed computing systems
Operating Strato involves running your own Linux virtual machine and learning how to use the Linux terminal environment. It can also be useful to understand secure connections through SSH and secure file transfer protocols.
Learn the Key Features
Discover the essential features of Strato.
-
Customizable Virtual Machines
Strato enables users to create and manage virtual machines, adjusting resources such as CPU, RAM, and storage to fit their specific research needs.
-
Scalable Computing Power
With Strato, you can scale computing resources up or down depending on the intensity of your workload, ensuring optimal use of resources and cost-efficiency.
-
CPU & GPU Support
Utilize powerful CPU and GPU resources for tasks like machine learning model training and simulations of complex systems.
-
Network Storage
Provides substantial network storage capabilities, allowing researchers to store and process large datasets directly within the platform without the need for external storage solutions.
-
Self-managed Environment
Strato offers a Linux-based environment where users have full control over their virtual machines, ensuring the ability to install and manage necessary software.
-
User Flexibility
Whether for data-driven research, minor services, or experimentation, Strato provides a personal server environment that users can customize and adapt to their research needs.
Use cases
Find out how Strato can be effectively utilized in certain computing practices.
-
Synthetic data generation and probabilistic graphical model
Integrate the joint power of CPU and GPU in an interactive environment provided by Strato if you need to run your experiments on servers with many CPU cores, for example for creating synthetic data. For some methods, just like probabilistic graphical models, you can take advantage of the GPU available on this system and run the algorithms effectively.
-
Developing natural language processing pipelines
Develop a natural language processing pipeline and create a set consisting of several natural language processing large language models. Involve extensive datasets in your projects with convenient memory and storage access possibilities of Strato with the possibility of GPU integration. Work in an individual and interactive environment.
-
Graph data management and knowledge graphs
If you are planning to carry out CPU-intensive processes in your project, consider integrating Strato, which allows for explorative research thanks to its interactive environment - for example running queries or specialized algorithms while working on the graph data. Suitable for those whose analysis involves measuring the efficiency of the code, and where the code algorithms play a great role in the project.
-
Knowledge graph embedding
Involve the powerful GPU provided by AI Cloud in your project to train different knowledge graph embedding models for many applications like entity recognition, clustering, or link prediction. Take advantage of parallel computing possibilities and integrate them in multi-relation learning of your models.
-
Text mining
Tackle multi-dimensional problems involving text mining methods and consult utilizing Strato, which provides access to powerful CPU cores. Set up your own virtual environment, establish the connection between the files, settings and libraries, and work on your extensive data sets at any time, without the need of setting up the environment every time you start running the virtual machine and save time on training your models.
-
Data mining
If your project involves dealing with a great amount of unstructured data, consider integrating Strato buy-in for the implementation of various mathematical and statistical algorithms in an interactive environment allowing for fast and parallel computing. Involve multiple complex databases and data processing and choose the required hardware specifications yourself.
-
Training object detection model
Train object detection models with the use of deep neural networks by integrating GPU in an explorative and interactive environment. The possibilities of Strato allow you to run tasks on servers with powerful GPU to integrate training pipelines, and effectively train your models.
-
Running parallel simulations
If you are planning to run several simulations in your project simultaneously, possibly also involving extensive data sets, you might consider applying for access to Strato, which allows for parallel computing in an interactive environment of virtual machines.
Important Information
Not for confidential or sensitive data
With Strato you are only allowed to work with public or internal information according to AAU’s data classification model (classified as levels 0 and 1, respectively).
If you would like to work with confidential or sensitive data (classified as levels 2 and 3), then we support another HPC platform called UCloud.
Review the terms and conditions
Before getting started, take a few moments to review the terms and conditions of using Strato, and don't hesitate to reach out to our support team if you have any questions or concerns.