Laboratories

Optimization of electricity usage in large-scale computing, assuming such use of the existing computing infrastructure to minimize electricity consumption, while not significantly affecting the efficiency of calculations and the quality level of the services provided.

The laboratory consists of 16 computing nodes, located in a server cabinet and connected by an Ethernet 1Gb / s network. During computational experiments, electricity consumption is monitored on several levels: power strips (PDUs), power supplies, and directly processors (CPU) and memory (RAM). The figure below shows a schematic of the laboratory:

Computing innovation on supercomputers in a wide range of reducing electricity consumption, e.g.:

1. Development of methods and development of software for multi-criteria optimization (energy, performance) for High Performance Computing (HPC) for CPU and GPU computing units, using hardware power-cap mechanisms.
2. Construction of electricity consumption models for the analysis of large-scale data sets (Big Data) using multi-node computer systems, with particular emphasis on modern data processing platforms (eg Apache Spark), using the Green Lab, created at CI TASK.

Development of mechanisms for the implementation of service environments increasing the functionality of computing cloud.

Enriching computing cloud with additional mechanisms related to availability, virtualization, and containerization.

Flexible adaptation to the user’s needs.

Analysis of the possibility of using the OCP (The Open Compute Project) standard to systematically enrich the computing infrastructure and exchange data with new elements while ensuring the compatibility of the calculations performed.

Creating a model infrastructure for assessing the suitability of the OCP standard and assessing the difficulty of its adaptation with the use of technologically different components. This applies to the time and cost of developing, upgrading, (updating) or transferring software.

Friendly extension of the computing infrastructure for the dynamically expanding needs of large-scale computing, including artificial intelligence and machine learning.

Creation of digital spaces enabling automatic data collection.

Activation of the network (LoRaWAN TASK) in the Tri-City and a part of Gdańsk Bay as an access infrastructure for various IoT devices and testing the quality of operation of such devices and the stability of transmission under various conditions of wave propagation and at different distances of these devices from individual network nodes. The nodes of the network are located in two stationary points, and IoT devices are located on mobile vehicles: a sailboat or a controlled flying object (pseudo drone). An example of the route of the conducted research is given below:

Exemplary locations of IoT devices: used sailboat and pseudo-drone

Registered users can use such a network and attach their own devices sending specific measurement data, which are then collected in a database, analyzed and visualized accordingly. This leads to new methods (outsourcing) of monitoring the selected environment.

Cooperation in the field of advanced computing in Europe under the Operational Program: Intelligent Development 2014-2020, Project No. POIR.04.02.00-00-B001/18

The entire project will be developed within the following laboratories:

• HPC and cloud processing laboratory (processing infrastructure),
• Laboratory of access to this processing infrastructure,
• Processing infrastructure management and monitoring laboratory,
• Archiving and long-term storage of research results laboratory,
• Distributed data management and transparent access to data laboratory,
• Processing infrastructure security laboratory.

National Platform for Integration of Research Infrastructures with Innovation Ecosystems under the Intelligent Development Operational Program 2014-2020 (Priority IV): Increasing the scientific and research potential, Measure 4.2. Development of modern research infrastructure for the science sector. The project will be implemented by 21 partners, the leader is the Poznań Supercomputing and Networking Center of the IChB PAN. The aim of the project is to build laboratories, then create innovative services (e.g. time and frequency) for the scientific community and business on their basis, and then their commercialization.

• Laboratory of the Municipal Computer Network (including renewal of the TASK network),
• Time and frequency standard laboratory,
• Smart Campus Laboratory as an element of Smart City – construction of a demonstrator of technologies used in a smart city, i.e. covering the entire campus with a “radio umbrella” enabling wireless data acquisition from sensors, campus video monitoring, monitoring and control of thermal energy, campus traffic monitoring, lighting control,
• Regional Living Laboratories of Innovation inspired by ICT,
• Cloud Services Laboratory,
• Laboratory of Multiscale Simulations,
• Laboratory and e-training services (in the field of PIONIER-LAB and innovation inspired by technology),
• Pre-incubator services – virtual office and shared services (legal and accounting), group work, telementoring.