Advanced Computing Laboratory (ACL)

• We have developed a family of methods and algorithms for static and dynamic planning of expert computation processes in cloud computing environments. The methods are based on the coevolutionary approach allowing not only to control characteristics of computational tasks but also to configure resources for their execution.
• Our staff have developed a domain specific language (DSL) that allows to describe computational tasks in terms of main constructs, reflecting features of computational processes in distributed environment. A translator has been developed for this language that allows to create a control script for VLUC based on text or graphic notation of composite application.
• We have developed a system of multi-scale models based on complex networks for various emergency situations: epidemics, crime and riots, gossip manipulation, evacuation during disasters. The models are represented in the form of composite applications in DSL and are suitable for usage within VLUC for generating computation scenarios.
• We have created the Early Warning System (EWS) and a system supporting decision making processes, for preventing floods.
• We enhanced EWS systems aimed at supporting decision making in emergency situations (terror attacks, criminal actions, panic during fires and floods, onset of epidemics etc.).
• We have implemented decision making systems based on VLUC for controlling epidemiological state and social tension (using data from social networks).
• Our Laboratory has conducted research of social processes based on formalism of complex networks whose characteristics are determined by the subject domain.
• We have developed a multi-scale model of propagation of information processes in complex networks and parallel algorithms implementing it that allow effective scaling to up to 100 hundred cores. The model is applicable for describing epidemiological processes in nonhomogeneous stratified population.

Education and career development:

• We have launched several programs in English language for «Applied Mathematics and Information Science» majors: «Supercomputer technologies in interdisciplinary research» (2012), «Urban Supercomputing», «BigData and Urgent Computing» (2014).
• More than 150 students have enrolled to education programs of the department highly productive computations of the St Petersburg University of Information Technologies, Mechanics and Optics - ITMO. The students represent Russia, ex-USSR countries, USA, Iraq, Kenya, Morocco, Nigeria, Ethiopia, France, Germany, as well as BRICS countries.

Organizational and structural changes:

We have launched the Center for Situational Modeling and Visualization (2011). As the core of the Center we have created a system of six Roxar 3D Projection design devices with a joint control system that projects images onto a reverse projection screen that interacts with audio and video systems, subsystems of video communication, computation and control. On the basis of the System we have conducted projects in modeling behaviors of marine vessels at various levels of roll, flights of aircraft and their landing a vessel situated at sea, problems of preventing floods and many other projects.

Other results:

2015 – Invention No 2569568. Bukhanovskiy A.V., Kovalchyuk S.V., Ivanov S.V., Nechayev Yu.I., Sloot P. «A method for controlling emergency situations based on integrating computational and informational components of a grid systems».