Novian Technologies has implemented a unique architecture of high-performance computing devices at Vilnius University, Faculty of Mathematics and Informatics (VU MIF), which enables fast and efficient research in the fields of artificial intelligence development, neural network training, and other creative and research activities that require very high scale calculations.
Situation
Researchers at one of the oldest high schools in Central and Eastern Europe, Vilnius University have been using high-performance computing devices for more than ten years. Technology has developed rapidly over the last decade, and research and its applications around the world have changed accordingly. Previously implemented technologies have limited the university community’s ability to compete effectively internationally. Thus, it was imperative for the university to fundamentally upgrade its existing infrastructure.
Problem
Most of the equipment produced, when used alone, is dedicated to relatively narrow tasks. No one such device can perfectly meet all the needs of a separate organization. A growing global practice is to design clusters of this equipment in a way to meet the specific complex needs of a particular organization best. The main benefit of such collections of equipment is that they allow to efficiently and quickly obtain the required tool in the face of a specific complex computational task. Besides, the specific architectural complexity and completeness of such a cluster provide speed and efficiency.
Solution
A cluster has been designed
High-performance computing systems and their components are currently manufactured by many global technology companies. In recent years, high-performance computing (HPC) clusters have been upgraded and their computing performance increased by using graphics processing units, and NVIDIA is the market leader in this segment production.
For the physical (technological) part of the solution, Novian Technologies (previously – BAIP) architects designed a complex high-performance computing cluster consisting of 1,728 units of non-accelerated computing cores from Intel and 20,480 units of computing cores from NVIDIA manufacturer Tensor.
The computing cluster has more than 13TB of RAM. The total calculated computational efficiency of this HPC cluster reaches 361 TeraFLOP (floating-point operations per second) DP (dual precision) and 4 PetaFLOP computational efficiencies in the field of deep learning. Thus, the computational performance of the current cluster TeraFLOP has increased about 12 times.
Data storage solution
A two-tier data storage solution based on the Luster ZFS file system has been designed for data storage. The control and data module has about 20TB of fast, SSD-based useful capacity, and all other data has about 0.8PB of useful capacity.
Computing and data storage resources are combined using InfiniBand EDR, 100Gb / s high-speed, low-latency networking technology. This allows computing and data storage resources to exchange information 2 times faster than in the previous VU MIF computing cluster.
For optimal use of resources and integration of the designed physical (technological) part with other HPC clusters, service management, self-service, and connection software module are installed. The software solution ensures the possibility to connect all high-performance computing nodes both with VU-owned and external, non-VU-owned, other high-performance nodes.
HPC nodes are combined
The implemented service management, self-service, and integration software module allows Vilnius University to combine the available resources HPC into one common computing resource, as well as to participate in EuroHPC and similar projects.
In parallel, a business plan, service catalog, and system accounting module are prepared for the university. Thus, it is possible to offer the market a sufficiently wide package of different technologies, easily adapt it to the customer’s needs, promptly prepare the required environment and then implements its accounting. It is expected that the opportunities provided by the renewed and modernized VU HPC cluster will be successfully used by both the university community and the country’s advanced businesses.
Result
During the project, Novian Technologies implemented and integrated state-of-the-art technologies from global manufacturers: the proposed combination of technologies from different manufacturers allows for efficient use of computing resources and rapid selection of technologies to solve specific tasks.
The upgraded VU HPC cluster is at least 10 times more efficient than the previous one. It will be used to model electronic systems for light energy harvesting and photosynthetic molecular electronic materials using quantum chemical methods. The ground and excited states of photosynthetic compounds will be analyzed, Raman spectra, the dynamical characteristics of the excited states, graphene layers and the possibility of proton transfer in conjugated systems will be modeled.
Organic materials and devices are expected to be widely used in fields ranging from medicine to energy and computing. Using a more powerful HPC system, VU researchers will be able to model quantum computers based on molecular compounds and develop quantum computing algorithms. In addition to VU research projects, the Open Access Centre’s HPC infrastructure is available to researchers from other Lithuanian research institutions and companies.
On the basis of these HPC resources and work, Lithuania has become a member of the EuroHPC project, and Vilnius University has become Lithuania’s representative in the joint EuroHPC Centre of Excellence for High Performance Computing. Membership of the EGI (Europe Grid Infrastructure – European supercomputing network) is also planned.
