- 1.2. – 30.4.2021
- Brno – hlavně pro studenty od třetího ročníku univerzitního studia
- Každé z témat je vedeno agilní metodou našim interním mentorem
- Hledáme vždy 2 studenty pro jedno téma
- Dlouhodobá spolupráce vítaná
- Stáže je placená hodinově
- Všechny témata jsou součástí reálných potřeb projektů
- Nepočítáme, že budeš všechny technologie znát do hloubky
- Angličtina na konverzační úrovni
Every system produces huge amount of logs which must be later analyzed (often manually) to discover source of the problem. This task is time consuming and the output of analysis can be also misleading. Therefore it is ideal task for automation. Machine Learning (ML) approaches allow computers to do it instead of humans. At first phase the system is trained using previously analyzed logs. In the second phase the ML system analyzes logs by itself and infers where is the problem.
Goal: Automatic data collection. Data filtration and anonymization. ML model design and implementation. Model learning. Deployment to the CI environment
- Basics of data analysis and statistics
- General machine learning knowledge
- NLP (Natural Language Processing)
- Python 3
- Knowledge about ML/Deep learning frameworks (TensorFlow, Keras, PyTorch, ...)
The subject of the project is to enhance functionality of the existing Java EE WEB application running under the WEB application server. External devices are physically interconnected and create networking fabric. The configuration information itself contains LLDP records based on actual state of interconnection.
Goal: The application will read network configuration information from several preconfigured external devices (network switches) via the REST interface and save it to its own database. User interface will show graphical scheme and topology of individual links interconnecting the devices including selected parameters.
- WEB application: Java EE (version 8+), PostgreSQL, Spring, Hibernation, Bootstrap
- REST, JSON
- Application Environment: Linux and WebLogic (Oracle) or WildFly (RedHat)
- Development environment: Maven, GIT, IntelliJ or Eclipse.
Docker vs Podman – comparison of container tools
Prerequisites - Get familiar with: -Lab documentation -Creation of new VMs in lab environment -Containerization -Ansible Steps -Prepare virtual environment for testing -Install Docker and test its features (same with Podman) -Use ansible for automation of basic Docker operation (same with Podman) -Compare their python libraries -Analyze docker usage in kolla-ansible and find the way how to replace it by Podman -Analyze the results and prepare documentation
The goal of the project is to analyze these two tools and compare their features in several aspects -Installation and configuration -Building of images -Basic usage -Automation with Ansible -Python libraries (Docker-py vs Python-podman) -Kolla-ansible (Is it possible to replace Docker by Podman?)
- Linux (CentOS) – configuration and scripting
- Basic knowledge of virtualization
- (optional) Docker
- (optional) Ansible
Kubernetes up and running
We’re going to install k8s on real hardware in our Lab. Also we have to automate as much as possible in terms of HW provisioning, VMs provisioning and k8s deployment itself. If we have enough time we can continue with making some app cloud-ready and deploy it on our cluster.
The main goal is to gain knowledge of top level k8s architecture, main k8s conceptions and abstraction, find out common ways to deploy k8s cluster and document findings.
- Linux CLI
- virtualization basics