Title

Surname

Name

Institute

Status

Mr

Kekelia

Besik

Georgian Technical U ...

presenter

Title: Investigation of Geometry Descriptions of End-Cap Toroid Magnet for ATLAS Simulation Software Infrastructure In parallel of the real experiment, simulation process of real experiment is underway at CERN. As a result two kind of data are received: 1. data coming from the real experiment 2. Data coming from the simulation process. In most cases they do not match each other’s. Several reasons can cause discrepancies between Data and Monte-Carlo. Several investigations show that they are coming from geometry descriptions in simulation. For the investigation of Data vs Monte-Carlo difference, investigation of ECT was carried out. For this purpose 3D model of ECT was build and compared to simulated ECT geometry.

Mr

Shekiladze

Davit

Georgian Technical U ...

presenter

GEOMETRY CREATION METHODOLOGIES FOR LHC POINT 4 Some geometries of LHC Point 4 were created with wrong ways, for example: we can't see parameters of geometries, can't change shapes and some of them are not finished. In some cases there are no geometry at all. After the investigation of this problem we corrected corrupted models and created missing ones. 26 sketches are corrected. Missing models are created on base of scanned geometry and pictures of real Point 4.

Mr

Udzilauri

Nikoloz

Georgian Technical U ...

presenter

The concept of ATLAS tracer Georgian engineering team is developing web Based interactive learning tool for ATLAS detector called "ATLAS tracer". Application running inside users browsers without any installations and suitable for all type of hardware and software platforms. ATLAS tracer provides event visualization and detailed as-built geometry representation of ATLAS detector, realistic rendering, zooming, rotation, movement, fly mode, cuts and measurements. So users can work like in professional CAD applications and learn detector deeply component-by-component.

Mr

Varamashvili

Davit

Georgian Technical U ...

presenter

## ATLAS Tracer engine and functionalities:

ATLAS Tracer is using WebGL based library, three.js . Which gives opportunity to display 3D objects in web browser. WebGL (Web Graphics Library) is a JavaScript API for rendering interactive 3D computer graphics and 2D graphics within any compatible web browser without the use of plug-ins All Major browsers (Chrome,FireFox,Opera,Safari,...) have WebGL support. ATLAS Tracer can be launched in any popular OS (Windows,Linux,Mac OS,Android,iOS)

Mr

Surmava

Archil

Georgian Technical U ...

presenter

DEVELOPMENT OF ANALYSIS AND SYNTHESIS METHODS OF GEOMETRICAL MODEL OF ATLAS DETECTOR ATLAS detector represent big and complex engineering construction. Engineering data is saved and managed by EDM system that contains different type assemblies and subassemblies. Consequently, development of hierarchical tree of ATLAS is necessary in order to build whole geometry model, which in turn contains classification and decomposition of atlas components. Formation of hierarchical tree was carried out with functionality, spatial and symmetrical decomposition of ATLAS detector.

Mr

Kobakhidze

Shota

Georgian Technical U ...

presenter

Development of Simplified Geometry Descriptions of ATLAS Detector for Tracer Software Application ATLAS Tracer is software application which uses data from ATLAS detector to visualize events such as displaying jets, electrons, tracks, muons etc. and gives information about each of them. During creation of Atlas tracer we stumbled upon a problem of geometries being too heavy for browser in which the program runs, to load them all together, they had too many faces(flat surface that forms part of the boundary of a solid object) which forced us to find a way to trim down amount of faces by simplifying given geometries.

Mr

Tsutskiridze

Niko

Georgian Technical U ...

presenter

INVESTIGATION OF GEOMETRY MODELING METHOD FOR ATLAS SIMULATION High Energy Physics (HEP) implements simulation for deep and wide range investigation of physics experiments by generating artificial events from the Monte-Carlo (MC) events generators in a format which is identical to the output of the facilities for experiment, detectors. However in some regions of detector data_vs_MC never match perfectly and there are discrepancies. Several reasons can cause it. Primary interest falling down on the investigations to understand how correct is detector representation in simulation. Simulation infrastructure implements GEANT for modelling of geometry. Shapes consistency and detalization is not important while adequateness of volumes and weight of detector components are essential for tracking. Geometry descriptions stored in database in general formats – XML, Oracle tables, etc. Then it is transforming in GEANT during the simulation. There are 2 main reasons of faults of geometry descriptions in simulation: 1. Difference between simulated and as-built geometry descriptions 2. Internal inaccuracies of geometry transformations added by simulation software infrastructure itself. For investigation of both reasons CATIA-GEANT interface was built. The aim was to use CATIA as a hub for collection of different geometry descriptions using by simulation and in same time make their comparison to find faults and analyse quality of GEANT simulation infrastructure.

Mr

Pataridze

Lasha

Georgian Technical U ...

presenter

EVENT VISUALIZATION FOR ATLAS Tracer ATLAS detector collects results of each collision and then stores this huge portions of the information into data base, to make it understandable for people, this information is described in different formats, like XML, but it's still hard for everyone to read those files, so in this case we have programs, like ATLAS Tracer. ATLAS Tracer is event display which gives opportunity to see the result of the proton proton collision using parameters described in XML files. Tracer visualizes particles directions, trajectories and also gives information about each of them.

Dr

Kvaratskhelia

Vakhtang

Muskhelishvili Insti ...

presenter

George Giorgobiani, Vakhtang Kvaratskhelia,Marine Menteshashvili On some applications of Hadamard matrices Abstract In the presentation a short survey of the theory of Hadamard (Sylvester) matrices is given. Moreover, numerical characteristic of Sylvester matrices is introduced and some of its properties are shown.

Dr

Menteshashvili

Marine

Muskhelishvili Insti ...

presenter

Dr

Sanikidze

Zaza

Muskhelishvili Insti ...

presenter

Dr.

Giorgobiani

George

Muskhelishvili Insti ...

participant

Dr.

Razmadze

Marina

Muskhelishvili Insti ...

participant

Mr

Oikashvili

Bekar

Muskhelishvili Insti ...

presenter

Modern supercomputers, Georgian supercomputer project and usage areas. Author: Bekar Oikashvili Co-authors: Prof. Archil Prangishvili, Prof. Zurab Modebadze Supercomputer industry and it's usage is growing every day in many developed countries, companies or institutes. More and more new tasks are executed and many problems are solved using these supercomputer platforms. Today Georgia is getting closer to those technologies. We are building first Georgian Supercomputer in Tbilisi, in Muskhelishvili Institute of Computational Mathematics of GTU. Primary missions for this supercomputer are to connect it to CERN Tier 3 network and execute CERN Tasks, to use it in meteorology and accurate weather forecast on territory of Georgia, to provide HPC platform for in-house usage for physicists and mathematicians and to allow students to test and use the platform for their tasks and experiments.

Mr.

Iashvili

Levan

SAQORGGAZI

participant

Dr

Sharmazanashvili

Alexander

CERN

presenter

Title: ATLAS Condition DB Management on the Base of COOL Tag Browser Development of COOL Tag Browser is one of the working package Georgian Team is executing in the framework of collaboration agreement between ATLAS and Georgian Technical University. 7 releases was built during the 1.5 years before application goes to production folder. Application contains 4’500 javascript strings and was developed by 1.5FTE of Georgian Team. Application implement REST full service and use PL/SQL and CherryPy services installed on standard aiatlas.cern.ch servers. Main functionality include COOL navigation and Data retrieve and visualization.

Prof

Barberis

Dario

University and INFN, ...

presenter

Title: Challenges in information technology for HEP Abstract: The LHC experiments successfully commissioned their software and computing infrastructures to support the physics program during LHC Run 2. The next phases of the accelerator upgrade will present new challenges in the offline area. In particular, at High Luminosity LHC (also known as Run 4) the data taking conditions will be very demanding in terms of computing resources: between 5 and 10 kHz of event rate from the high-level trigger system to be reconstructed (and possibly further reprocessed) with an average pile-up of up to 200 events per collision and an equivalent number of simulated samples to be produced. The same parameters for the current run are lower by up to an order of magnitude. While processing and storage resources would need to scale accordingly, the funding situation allows one at best to consider a flat budget over the next few years for offline computing needs. In this paper we present a study quantifying the challenge in terms of computing resources for HL-LHC and present ideas about the possible evolution of the computing models, the distributed computing tools, and the offline software to cope with such a challenge.

Dr.

Modebadze

Zurab

Parliament of Georgia

presenter

Mr

Meladze

Hamlet

Muskhelishvili Insti ...

presenter

On One Nonlocal Contact Problem for Elliptic Equation and its Numerical Solution Hamlet MELADZE, Tinatin DAVITASHVILI The present work is devoted to the specific nonlocal statement and analysis of a contact problem for Poisson’s equation in two-dimensional domain. For numerical solution the iteration process is constructed, which allows one to reduce the solution of the initial problem to the solution of a sequence of classical Dirichlet problems. The algorithm is suitable for parallel realization. The specific example is considered and solved numerically.

Mrs

Davitashvili

Tinatin

Ivane Javakhishvili ...

presenter

Mr

Di Girolamo

Alessandro

CERN

presenter

Evolution of Distributed Computing for the ATLAS experiment at CERN ATLAS Distributed Computing during LHC Run-1 was challenged by steadily increasing computing, storage and network requirements. In addition, the complexity of processing task workflows and their associated data management requirements led to a new paradigm in the ATLAS computing model for Run-2, accompanied by extensive evolution and redesign of the workflow and data management systems. The new systems were put into production at the end of 2014, and gained robustness and maturity during 2015 data taking. ProdSys2, the new request and task interface; JEDI, the dynamic job execution engine developed as an extension to PanDA; and Rucio, the new data management system, form the core of the Run-2 ATLAS distributed computing engine. One of the big changes for Run-2 was the adoption of the Derivation Framework, which moves the chaotic CPU and data intensive part of the user analysis into the centrally organized train production, delivering derived AOD datasets to user groups for final analysis. The effectiveness of the new model was demonstrated through the delivery of analysis datasets to users just one week after data taking, by completing the calibration loop, Tier-0 processing and train production steps promptly. The great flexibility of the new system also makes it possible to execute part of the Tier-0 processing on the grid when Tier-0 resources experience a backlog during high data-taking periods. The introduction of the data lifetime model, where each dataset is assigned a finite lifetime (with extensions possible for frequently accessed data), was made possible by Rucio. Thanks to this the storage crises experienced in Run-1 have not reappeared during Run-2. In addition, the distinction between Tier-1 and Tier-2 disk storage, now largely artificial given the quality of Tier-2 resources and their networking, has been removed through the introduction of dynamic ATLAS clouds that group the storage endpoint nucleus and its close-by execution satellite sites. All stable ATLAS sites are now able to store unique or primary copies of the datasets. ATLAS Distributed Computing is further evolving to speed up request processing by introducing network awareness, using machine learning and optimization of the latencies during the execution of the full chain of tasks. The Event Service, a new workflow and job execution engine, is designed around check-pointing at the level of event processing to use opportunistic resources more efficiently.

doctor

Pkhovelishvili

Merab

GTU Muskhelishvili ICM

presenter

Natela Archvadze, Merab Pkhovelishvili, Lia Shetsiruli, Otar Ioseliani. The algorithm of parallel programming using “small delay”. In the article is reviewed a new effective algorithm working principal for sorting using parallel programming. Dividing of initial array is performing by the principal of “distribution of cards” . First element is being transferred to first core the second element to second core until n element (n is a number of cores ) to n-th core. The element n+1 is not only transferred to first core its being executed the operation of sorting as well with the elements located there so far. The n+2 element is being transferred analogically to second core and its being executed sorting the elements located there. The process is continuing until all of array elements will not be split into sorted separately n subarrays. The traditional merge operation is being executed after. The main win of time happens because sorting of arrays starts not after collapse arrays it starts in parallel with collapse. Each core starts sorting operation in parallel with “small delay” in compare with previous core, with received each new element and located there elements of already sorted array.

Dr

Mamporia

Badri

Muskhelishvili insti ...

presenter

On modeling of the turbulent movement Badri Mamporia Affiliation: Niko Muskhelishvili Institute of Computational Mathematics, Technical University of Georgia, Tbilisi, email: badrimamporia@yahoo.com Abstract. On investigation of the turbulent movement, the main problem is to give the velocity as the random process at any fixed point of the environment. According to our approach, the quantity of pulsations of the velocity at any fixed point is a Poison process; the random process of the pulsation part of the velocity is a Levy process. It follows that the velocity of the turbulent movement is a random element in suitable functional space. The theoretical achievements in development of the turbulent movement and experimental data are the foundations to give the covariance operator of this random element and the character of the randomness on the linear functionals of this random element. After preparing these necessity we can consider the corresponding stochastic differential equation of the trajectory in turbulent environment. The solution of this equation is the problem of future developments. Another problems are to give the approximative solution as a random process of this equation and to simulate in a computer the random process for various covariance operators to find the real, close to the objective account. Current advance of the computation technology and promised to our Institute high- capacity cluster give birth to hopes to receive an acceptable model of the turbulent movement.

Mr

Berikashvili

Valeri

Muskhelishvili Insti ...

participant

Dr

Papiashvili

Maguli

Muskhelishvili Insti ...

presenter

No

Abramidze

Edisoni

N. Muskhelishvili In ...

participant

Dr

Nachkebia

Mziana

GTU Musxelishvili in ...

presenter

Dr.

Zarnadze

David

GTU Muskhelishvili I ...

presenter

On calculation of the inverse of multidimensional harmonic oscillator on Schwartz space D.N.Zarnadze, D.K.Ugulava, M.G.Kublashvili, P.Tsereteli Abstract Classical harmonic oscillator operator defined on Schwartz spaces of rapidly decreasing functions on one dimensional Euclidean space is largely used in quantum mechanics for white noise analysis. In the paper [1] the least squares method for approximate calculation of the inverse of classical harmonic oscillator was used for approximate solution of the corresponding inhomogeneous equation. The convergence and some estimates of convergence of a sequence of the approximate solutions to the exact solution was proved. In this report p-dimensional analogy of classical harmonic oscillator operator A is defined on square integrable function space on p-dimensional Euclidean space. For this operator A the space of all orbits orb(A,x) is constructed [2] and special sequence of Hilbertian norms is considered. The isomorphism of this space and the Schwartz space of rapidly decreasing functions is proved. The least squares method [3]-[4] for approximate calculation of the inverse of p-dimensional harmonic oscillator in Schwartz space is used. As a basic functions the multiplied Hermits functions ([5], p.491) is considered. The convergence and some estimates of convergence of a sequence of a approximate solutions to the exact solutions is proved. The centrality (strongly optimality) [6] of this algorithm is proved. As well method for calculation of best approximation quantity with respect to the metric on the Schwartz space is given. Numerical experiments is also conducted that confirm the received theoretical results and shown necessity of quick and parallel computing. References 1. Zarnadze D.N. and Ugulava D.K. The least squares method for harmonic oscillator operator in Schwartz space. Intern. Conf. on comp. science and appl. Math., Conference’s Proceedings, March 21-23, 2015, Tbilisi, Georgia, 255-261. 2. Zarnadze D.N. and Tsotniashvili S.A. Selfadjoint operators and generalized central algorithms in Frechet spaces. Georgian Mathematical Journal, V 13 (2006), N 2. p 1-20. 3. Zarnadze D.N. A generalization of the method of least squares for operator equations in some Frechet spaces. Izv. Akad. Nauk Russia. Ser.Mat. 59 (1995). 59-72; English transl. in Russian Acad. Sci Izv. Math. 59:5 (1995), 935-948. 4. Zarnadze D.N. On the generalization of the least squares Method for the operator equations in some Frechet spaces. Int. Symposium “Operator Equations and Numerical Analysis”. Sept. 1992. Preprint N 22, Berlin, 1992. 5. Michlin S.G. Presdorf S. Singular Integral Operators. Springer Verlag. Berlin Heidelberg New York Tokio. 6. Traub J.F., Wojniakowski H., Wasilkowski G. Information based Complexity, NY, Acad.Press.1986.

Dr.

Ugulava

Duglas

GTU Muskhelishvili I ...

presenter

Doctor

Nikoleishvili

Mikheil

N. Musskhelishvili I ...

presenter

Doctor

Tarieladze

Vaja

N. Muskhelishvili In ...

presenter

COMPUTATION AL ASPECTS OF A DISCRETE EXSTREMUM PROBLEM MIKHEIL NIKOLEISHVILI AND VAJA TARIELADZE ABSTRACT. Two problems related with a discrete extremum are formulated. Introduction In what follows, N={1,2…}, Z_(+ )= {0.1,2…} Let L ?1,n >1 be natural numbers and let B (L, n)={(x_1,…?,x?_n )?N^n ? ?_1^n?x_i =L} Moreover, we fix ( k_1, ….k_n) ??Z_+?^n and (x_1,…?,x?_n )??X_+?^n and consider the set B (L, n; k_1,…,k_n) = {(x_1,…?,x?_n )?B (L,n): ?,x?_i >k_i, i=1,…n} Clearly B (L, n; 0,…,0)= B (L, n) (1.1) We assume that L ??_1^n??k_i+? n (1.2) The condition (1.2) guarantees that B (L, n; k_1,…,k_n) ?? (1.3) Since (1.3) is satisfied and the set B (L, n; k_1,…,k_n) is finite, the extremum b(L, n; k_1,…,k_n; s_1, ….s_n) of the set {?_1^n??(x?_i +s_i): (x_1,…?,x?_n )?B (L,n;k_1,…,k_n ) } Is well-defined by the equality: b(L, n; k_1,…,k_n; s_1, ….s_n)= max{?_1^n??(x?_i +s_i): (x_1,…?,x?_n )?B (L,n;k_1,…,k_n ) } . -2- The exact formula for b(L, n; k_1,…,k_n; s_1, ….s_n) is known only in the following particular case: Proposition 2.1. ([1]) Let k ?0 and L, n be natural numbers such that L?n(k+1). Then b(L, n; k,…,k; 0; ….0) =?(1+q)?^r q^(n-r) , where q=[L/n] and r=L-nq In our communication we are going to discuss the following problems: Problem 2.2. Find a formula (or find the sharp estimations from below and from above) for the cardinality (number of elements) of the set B (L,n;k_1,…,k_n ) Problem 2.3. Find a general algorithm for calculating of b (b(L, n; k_1,…,k_n; s_1, ….s_n) and estimate its computational complexity. Acknowledgements. The authors were supported in part by Shota Rustaveli National Science Foundation grant no. FR/539/5-100/13. REFERENCES Mikheil Nikoleishvili, Vaja Tarieladze, About a problem of extremum, Proceedings of the Fifth International Scientific-Practical Conference “Scientific Issues of the Modernity” at Sukhishvili Teaching University, Gori, Georgia, December, 13, 2014, pp. 401-403. NIKO MUSKHELISHVILI INSTITUTE OF COMPUTATIONAL MATHEMATICS OF THE GEORGIAN TECHNICAL UNIVERSITY, 0175 TBILISI, GEORGIA E-mail address: mikheil.nikoleishvili@gmail.com NIKO MUSKHELISHVILI INSTITUTE OF COMPUTATIONAL MATHEMATICS OF THE GEORGIAN TECHNICAL UNIVERSITY, 0175 TBILISI, GEORGIA E-mail address: v.tarieladze@gtu.ge -2-

Prof.

Abuladze

Inga

Georgian Technical U ...

presenter

Concept of Cloud Computing Inga Abuladze Georgian Technical University E-mail: i_abuladze@gtu.ge Cloud computing is the result of the evolution and adoption of existing technologies and paradigms. The goal of cloud computing is to allow users to take benefit from all of these technologies. The cloud aims to cut costs, and helps the users focus on their core business instead of being impeded by IT obstacles. new concepts are mixed with new ones, and some old technologies are being reinvented. Today, many research areas and projects under the cloud umbrella need to be sorted and classified. Accordingly, this survey article presents a unified view of the Cloud Computing main concepts, characteristics, models, and architectures.

Mr

Bakradze

Giorgi

Productsavvy

participant

Mrs

Khalilova

Shahla

Institute of Physics

participant

Dr

Adam-Bourdarios

Claire

LAL

presenter

prof.

Asatiani

Pavle

Georgian Technical U ...

presenter

Circulation physical fundamentals of computing and technologies On the basis of discrete subplanckian unitary methodology of relations and interactions systems computing in the nature and society is proposed, together with new architecture of internet and machine language based on the circulation. Independently in the same direction the scientists of Massachusetts Technology Institute (USA) on the level of electron spin are engaged; in accordance with our paradigm the spin and Planck constant are the product and derivative of circulation. So we have come to the postplanckian fundamentals of computing.

Mr

Razmadze

David

Georgian Technical U ...

participant

Dr

Zakradze

Mamuli

Muskhelishvili Insti ...

presenter

ON SOLVING THE DIRICHLET GENERALIZED PROBLEM FOR A HARMONIC FUNCTION IN THE CASE OF AN INFINITE PLANE WITH A CRACK-TYPE CUT M. Zakradze, M. Kublashvili, Z. Sanikidze, N. Koblishvili Abstract. An algorithm for the approximate solution of the Dirichlet generalized problem is proposed. The term generalized indicates that boundary function has a finite number of first kind break points. The solution consists of the following stages: 1) The reduction of the Dirichlet generalized problem to an ordinary auxiliary problem for a harmonic function; 2) The approximate solution of the auxiliary problem by the modified version of the MFS (the method of fundamental solutions); 3) The construction of the approximate solution of the generalized problem from the solution of the auxiliary problem. An example is considered in which the break points are cusp points.

Dr

Jochen

Meyer

Nikhef

presenter

It is a common approach to compare simulated to actual data in order to search it for new phenomena or give interpretations in terms of predictive theories. This concept is also prevalent in modern high energy physics. The ATLAS experiment at the Large Hadron Collider at CERN is recording the outcome of high energetic particle collisions which are produce with a rate of 40 MHz. For statistically reasonable statements billions of such events are required to be simulated, which constitutes a tremendous technical effort. Therefore one body of the software project of the ATLAS experiment is dedicated to concerns of this topic only. In this presentation, several key aspects are highlighted. Details on the technical performance of the simulation software infrastructure will be given with a special focus on the interplay of the various programs used. Furthermore, comparisons of simulated data and actual data will be shown to demonstrate the importance of a realistic simulation. Finally, plans to improve above-mentioned subjects in the future will be pointed out.

Dr

Cheatham

Sue

Udine

presenter

Title : ATLAS outreach highlights Abstract : The ATLAS outreach team is very active, promoting particle physics to a broad range of audiences including physicists, general public, policy makers, students and teachers, and media. A selection of current outreach activities and new projects will be presented. Recent highlights include the new ATLAS public website and ATLAS Open Data, the very recent public release of 1 fb-1 of ATLAS data.

Mr

Martelli

Edoardo

CERN

presenter

The computer network is a fundamental part of the LHC instruments: the data produced by the Experiments have to be moved from the detectors to all the institutions around the world collaborating with CERN to be analysed and understood. The presentation will give an overview of the networks used inside CERN to connect the accelerator, the datacentres and all the users. As well, it will present the wide range networks used to distribute the LHC data to datacentres located all around the world.

Dr.

Giorgi

Ghlonti

Muskelishvili Instit ...

presenter

Design of a Cyber Infrastructure for Accumulation and Distribution of Analytical Information Resource The aim of the project is to develop an architectural solution for a cyber infrastructure intended for accepting, accumulation and distribution of analytical figures (from subject areas such as healthcare, education, production, etc…), their transformation into corporate analytical information resource, convenient for the purpose of decision making at different levels of organizational hierarchy, with ability for all stakeholders to gain access to it on the basis of their rights and (presumably) economic and civil relations. The corporate analytical information resource is regarded as a union of data clusters comprising analytical information resources of different subject areas. Solution involves, for each cluster, data acquisition, on some regular basis, from primary information providers of a subject area, use of Cloud infrastructure as a service for their further processing and transformation into information aggregates, relevant to various levels of subject area’s organizational hierarchy, maintenance of long-term information archives, and subsequent provision of information needs of different stakeholders Data collection according to coordinated models ensures integrity and consistency of information resource, comparability and compatibility of data, transparency of information space. Among requirements to cyberspace standard requirements of openness, security, scalability and elasticity might be mentioned, as well as responsibility for quality, integrity, transparency, timeless and low price of information resource accumulated. The users of cyber infrastructure are provided by different information services, both routine and ad-hoc. They are equipped by the tools allowing the specification of appropriate higher-order queries.

Dr.

Chelidze

George

Tbilisi State University

presenter

An algorithm for distributing jobs in cluster environment G.Chelidze, B. Mamporia, Nodar Vakhania Affiliations: Tbilisi State University, Niko Muskhelishvili Institute of Computational Mathematics of Georgian Technical University, Facultad de Ciencias, UAEMor Cuernavaca 62210, Mexico. E-mails g.chelidze@tsu.ge , badrimamporia@yahoo.com nodari@uaem.mx In cluster scheduling serial and parallel jobs need to be distributed among parallel CPUs. The corresponding combinatorial optimization problems which arise here are intractable (NP-hard). Hence one cannot expect to "solve" such problems optimally. Besides, the objective criteria are often contradictory (there might be no single criterion). Here we propose a heuristic method that can be used for the distribution of jobs on parallel CPUs with the objective is to minimize CPU idle-time. Our algorithm works on non-identical CPUs when the speed of a CPU is job-dependent.

Dr.

Julakidze

Levan

Georgian Technical U ...

presenter

"New Tweakable Block Cipher" Levan Julakidze, GTU Informatics and Control Systems Faculty, Doctoral degree in Informatics. Zurab Kochladze, Candidate of Technical Sciences, TSU Associated professor. Tinatin Kaishauri, Candidate of Technical Sciences, GTU Full professor. As it is widely known for protection of the information confidentiality generally symmetric block algorithms shall be applied, as the open key systems speed is quite low. Unfortunately, the block ciphers have significant fallback. That is their determined nature, which is expressed in the fact that the same text by means of the same keys is always transferred into the same ciphertext. This fallback is tried to be suppressed by means of the encrypting regimes, in which the initialization vector is applied for, which enables to transfer the same text with the same keys into various cipher texts. In 2002 the Article by M. Liskov, R. Rivest and D. Wagner was published, the idea stated in which, initialization vector might be used not in the regime of encrypting, but in the algorithm itself. Such ciphers are called tweakable block ciphers. In our work the problem of construction of such new cipher is overviewed by means of the method Hill. As it is known, Hill algorithm is one of the best methods to achieve diffusion, but it is the linear transformation, due to which this algorithm in modern cryptography was not applied. The algorithm block scheme developed in the work involves operations from various algebras, which are the guaranty, that the encrypting algorithm will not be linear, and it will be impossible to implement the algebra attack onto it.

Dr

Michael

Hoch

CERN/ CMS

presenter

- art@CMS sciART exhibitions and Workshops - art@CMS SciArt Workshops are interdisciplinary workshops designed to introduce school, college, and university art and science students to the scientific world of particle physics and fundamental research through artistic enquiry. They incorporate an introduction to particle physics, CERN and CMS, followed by a philosophical reflection and discussion session designed to inspire the students to create their own, science related artworks. The students then have the opportunity to mount an exhibition of their work. Forgoing more traditional educational models, students construct and develop their own non-declarative projects, i.e. projects which begin without having pre-defined outcomes. As such, students follow a design framework which promotes creativity, efficacy and most importantly, acts as a useful step towards autonomous learning. Working in such non-declarative and concept-based learning environments is far more demanding of the teacher as well as the learner. It is in order to respond to this added difficulty that the design framework was conceived; it is a flexible tool to help both teacher and learner navigate these complex waters. The advantage is obvious; teaching through concepts is a teaching that can facilitate understanding in a similar way to the way in which pictograms facilitate navigation. The design framework acts as a non-prescriptive scaffolding which the learner can use to support their concept from brainstorming, research, analysis, development, mock-up, final to post project analysis. We see this as a 360 degree design exercise and not just a linear one. SciArt Workshops foster skills that are more appropriate to the challenges of the 21st century. Much work has already been done by the likes of Prof. Mihaly Csikszentmihalyi to quantify the benefits of developing skills such as innovation, creativity and critical thinking. The excellent work of institutions such as the RSA show other examples of such work in action. Aiming to act like an ideas factory, SciArt Workshops are built on a platform of learning, sharing and sustainable development; a form of learner(s) hub where the individual can take ownership of their ideas, honing them through a rigorous learning process to the point where they can share them with others.

Dr

Storr

mick

university of birmingham

participant

Mr

Korzh

Vladimir

Muskhelishvili Inst

participant

Mr

Mkrtchyan

Tigran

ANSL (Yerevan Physic ...

presenter

The Tile Calorimeter (TileCal) is the central section of the ATLAS hadronic calorimeter at the Large Hadron Collider (LHC). The LHC has devised a series of upgrades towards a High Luminosity LHC (HL-LHC). A significant LHC luminosity upgrade requires increasing of calorimeter granularity because the current TileCal granularity will pose limitation for jet substructure, high boosted pT jets, jet energy resolution, etc. The proposed solution is the replacement of existing R5900 photomultiplier tubes by 8x8 multi-anode photomultipliers (MAPMT) without disassembling the module and keeping the Tilecal optics and mechanical structure intact. The use of MAPMTs is a progressing direction of the scintillation light detection, quite compact, with increased efficiency (app. 40% in the time being), tolerant to a magnetic field and expected very low cost in the near future. We have created a standalone test bench facility at CERN that allows us to implement new methods while maintaining an environment similar to the modules. The test bench measurements help us to understand the new read-out electronics, which differ from the ones that are currently used in the TileCal modules. Presently the TileCal cell structure is determined with fiber bundles coupled with PMTs which lead to certain fixed geometrical resolution. For each cell, the fiber bundle is unique and the fibers are distributed randomly. The main focus is directed to determine the optical properties of the coupling between the fiber bundles and the MAPMT, the expected geometry resolution and a way of possible improvement in the coupling.

Dr

Lange

David

Princeton University

presenter

The CMS offline software and computing system has successfully met the challenge of LHC Run 2. In this presentation, we will discuss how the entire system was improved in anticipation of increased trigger output rate, increased rate of pileup interactions and the evolution of computing technology. The primary goals behind these changes was to increase the flexibility of computing facilities where ever possible, as to increase our operational efficiency, and to decrease the computing resources needed to accomplish the primary offline computing workflows. These changes have resulted in a new approach to distributed computing in CMS for Run 2 and for the future as the LHC luminosity should continue to increase. I will discuss these changes, their impact on the CMS production and analysis computing environment and on-going work attempting to take advantage of new and evolving software development, analysis techniques and computing technologies.

The ...

Archvadze

Natela

Ivane Javakhishvili ...

presenter

Natela Archvadze, Merab Pkhovelishvili, Lia Shetsiruli, Otar Ioseliani. The algorithm of parallel programming using “small delay”. In the article is reviewed a new effective algorithm working principal for sorting using parallel programming. Dividing of initial array is performing by the principal of “distribution of cards” . First element is being transferred to first core the second element to second core until n element (n is a number of cores ) to n-th core. The element n+1 is not only transferred to first core its being executed the operation of sorting as well with the elements located there so far. The n+2 element is being transferred analogically to second core and its being executed sorting the elements located there. The process is continuing until all of array elements will not be split into sorted separately n subarrays. The traditional merge operation is being executed after. The main win of time happens because sorting of arrays starts not after collapse arrays it starts in parallel with collapse. Each core starts sorting operation in parallel with “small delay” in compare with previous core, with received each new element and located there elements of already sorted array.

Prof.

Tsereteli

Paata

St.Andrew the First- ...

participant

Prof

Janikashvili

Nona

Tbilisi State Medica ...

presenter

Mathematical modeling of immunopathogenesis of rheumatoid arthritis K.Odisharia, V.Odisharia, P.Tsereteli, N.Janikashvili Rheumatoid arthritis is a systemic autoimmune disease characterized by the joint inflammation and the cartilage destruction. Autoreactive B lymphocytes represent integral elements of the pathophysiology of rheumatoid arthritis. Immune balance between the effector and the regulatory T cell subsets guide the production of autoantibodies by B lymphocytes and, therefore, play a cardinal role in disease severity. While targeted therapeutic approaches are successfully emerging in medical practice, refined personalized analyses of T and B lymphocyte subsets in patients with rheumatoid arthritis are critically needed for rigorous disease management. Mathematical models of immune mediated disorders provide an analytic framework in which we can address specific questions concerning disease immune dynamics and the choice of treatment.Herein, we present a novel mathematical model that describes the immunopathogenesis of rheumatoid arthritis using non-linear differential equations. The model explores the functional dynamics of cartilage destruction during disease progression, in which a system of differential equations deciphers the interactions between autoreactive B lymphocytes and T helper cells. Immunomodulatory relation between pro-inflammatory and regulatory T lymphocyte subsets is also solved in these equations. Of importance, our model provides a mechanistic interpretation of targeted immunotherapy which deals with the intervention of pathophysiological immune processes in rheumatoid arthritis. In conclusion, we propose a novel mathematical model that best describes the immunopathogenic dynamics in patients with rheumatoid arthritis and, therefore, may take a rapid pace towards its implementation in biomedical and clinical research.

Prof.

Odisharia

Vladimer

Tbilisi State University

participant

Prof ...

Lomidze

Ilia

Georgian Technical u ...

presenter

Miss

Chachava

Natela

Georgian Technical U ...

presenter

Symbolic estimation of distances between eigenvalues of Hermitian operator Ilia Lomidze, Natela Chachava (Georgian Technical University, Tbilisi, Georgia) E-mail: lomiltsu@gmail.com, chachava.natela@yahoo.com Abstract We find out a method for symbolic estimation of a minimal (maximal) distance between eigenvalues of a Hermitian matrix (or roots of a polynomial with real (maybe degenerated) roots), using Hankel matrices formalism. The range of location of eigenvalues is symbolically estimated too. All estimations can be done with any precision.

Dr

Chobanyan

Sergei

Muskhelishvli Instit ...

presenter

A Monte-Carlo algorithm for finding a near optimal rearrangement of the Steinitz functional L.A. Chobanyan Institute TECHINFORMI, Georgian Technical University S.A. Chobanyan Muskhelishvili Institute of Computational Mathematics, Georgian Technical University Abstract Many problems of machine learning [1,2], scheduling theory [3,4,5] and discrepancy theory [2] lead to the problem of finding the minimum in permutations of the Steinitz functional. We suggest a Monte-Carlo algorithm for finding a near optimal permutation of the Steinitz functional. The construction uses the transference lemma [4] and the tail distribution of the Bernoulli random variable. Literature 1. N.Harvey, S.Samedi, Near-Optimal Herding, Workshop and Conference Proceedings,35, 1-18, 2014. 2. J.Beck,V.T.Sos, Handbook of Combinatorics, v. 2, Elsevier Science B.V. and the MIT Press, 1995. 3. N. Makai, Reroute sequence planning in telecommunication networks and compact vector summation, Appl. Math. Comp, 150,2004, 785-801. 4. L.A.Chobanyan, S.A.Chobanyan, G.Giorgobiani, A maximum inequality for rearrangement of summands and its applications in orthogonal series and scheduling theory, Bul.Georgian Nat. Acad. Sci.,5,16-21, 2011. 5. L.A.Chobanyan, V.V. Kvaratskhelia, An Algorithmic Solution to the problem of compact vector summation with an applications in scheduling theory, 9th International Conference CSIT-2013, Yerevan, Armenia, Proceedings: pp. 58-60, 2013.

Dr

Chobanyan

Levon

Institute Techinform ...

participant

Dr

Astsatryan

Hrachya

National Academy of ...

presenter

Cloud computing technology represents a new paradigm for the provisioning of computing resources by shifting the location of resources to the network to reduce the costs associated with the management of hardware and software resources. The aim of the article to give a brief overview of the the National Research Cloud Environment of Armenia, as a key component of a national e-infrastructure.

Prof ...

Gevorkyan

Ashot

Institute for Inform ...

presenter

Modeling of spin glasses from the first principles of classical mechanics A. S. Gevorkyan and V. V. Sahakyan Institute for Informatics and Automation Problems, NAS of RA, Str. P. Sevak 1, Yerevan 0014, Armenia It is studied the classical 1D Heisenberg spin glasses assuming that spins are spatial. The system of recurrence equations is derived by minimization of the nearest-neighboring Hamiltonian in nodes of 1D lattice. We have proved that in each node of the lattice there is a probability that the solution of recurrence equations can bifurcate. This leads to the fact that, performing consecutive node-by-node calculations on the n-th step instead of a single stable spin-chain we get a set of spin-chains, which form Fibonacci subtree (graph). We have assessed the complexity of computation of one graph and have shown that it is ~2^nK_s, where n and K_s denote the subtree's height (the length of spin-chain) and the Kolmogorov's complexity of a string (the branch of subtree) respectively. It is shown that the statistical ensemble may be represented as a set of random graphs, where the computational complexity of each graph is NP hard. It is proved, that all strings of the ensemble have the same weights. The latter circumstance allows in the limit of statistical equilibrium with predetermined accuracy to reduce the NP hard problem to the P problem with complexity ~ NK_s, where N is the number of spin-chains in the ensemble. As it is shown by comparing statistical distributions of different parameters, which are performed by using NP and P algorithms the coincidence of the corresponding curves is ideal. This allows to claim that it is possible to calculate all parameters and the corresponding distributions of the statistical ensemble from the first principles of classical mechanics without using any additional considerations. Finally, using formal similarity between of the ergodic dynamical system and the ensemble of spin-chains, it is proposed a new representation for the partition function in the form of one-dimensional integral from the spin-chains' energy distribution.

The ...

Shetciruli

lia

Batumi Shota Rustave ...

participant

Mr

Avaliani

Giorgi

Leavingstone

participant

Mr.

Zarkua

Teodore

Georgian Technical U ...

participant

Mr.

Kobakhidze

Paata

Muskhelishvili Institute

participant

Mrs

Sargsyan

Laura

CERN/BE

presenter

Mr

Vakhtang

Vakhtangishvili

Leavingstone

presenter

Leavingstone is a digital creative agency colliding technology and bright ideas since 2010. The first advertising company in the region to win a Cannes Lion at the Cannes Festival of Creativity (twice in a row), and partner up with Google to bring precisely targeted ads to the Georgian Internet is focused on using cutting edge tech and layers of big data to solve human problems. The presentation by the co-founder of the agency will feature an in-depth analysis of award winning cases, and a few ongoing projects using technology as a medium to bring ideas and humans together.

Mrs

Mirziashvili

Ketevan

CERN

presenter

EXPRESSION METHODS IN Art@Science ABSTRACT European Organization for Nuclear research (CERN) is well known scientific school. However not everyone understand in depth what scientific and technology achievements are reached there. Among of many interesting projects at CERN there is Art@Science project, which aimed to interpret CERN achievements in artistic way and make it popular in different societies. Two general activities can be separated : e-painting and e-design. E-painting foresee getting images from different sources like sketches, photos, paper paintings, etc and proceed them - cleaning, recoloring, rescaling, etc in paint studio. E-design foresee composition scenes in design studio. Popular slogans can be interpreted by direct and indirect ways. Presentation contains some works represented on the the exhibition organized by CMS at CERN.

Dr

Sanikidze

Jemal

Muskhelishvili Insti ...

presenter

On construction and application of some quadrature formulas of high accuracy for Cauchy type singular integrals J. Sanikidze, M. Kublashvili The questions of construction of approximation formulas for Cauchy type singularity integrals have been considered. Applicability of the indicated quadrature formulas for numerical solution of problems of applied physics and mechanics has been studied.

Dr

Kublashvili

Murman

Georgian Technical U ...

presenter

Dr

Sahakyan

Vladimir

IIAP NAS RA

presenter

Some Scheduling Algorithms for Computational Grid Systems The optimal usage of CPU times in multi-core computational Grid infrastructures depend on several factors, such as the job scheduling, the possibility of dynamic allocation of computational resources, the possibility of job migration to the different phases of implementation in optimal requested environment, performing a stop of the job with the possibility of continuing, etc. The use of grid infrastructures to handle the information stream requires a flexible approach to the allocation of resources as well for the timely performance of jobs. The job receiving process in the queue for further executing in a grid infrastructure plays an important role in the organization of the whole process. At this stage the incoming job service script is generated based on a maintenance data bursts. This requires synchronization of distributed processes to manage the resources of a grid system. A set of GRAM (Grid Resource Allocation and Management) services allows to manage the resources and jobs, job queues handling, ensuring their execution on the required computational resources. In addition to GRAM services, the schedulers play special roles. The adoption of the job queue for a service applies to the scheduler responsible for ensuring its timely implementation. Nowadays the usage of modern schedulers, such as Maui, Condor, PBS (Portable Batch System) have opportunities to perform the planning tasks on specified priorities. However, in complex cases (for instance to run the job at a specified time and in specified resources, or to manage multi-stream queues) we need to implement flexible approach for scheduling. The optimal run of the programs in multi-core environments involve selection of service discipline, delivering at least a "loss function", which characterizes the quality of the functioning of the process. As such a function can be considered some functional depending on the average waiting time, utilization of time, the number of denial of service and others. Many authors have shown that the discipline of the relative priority is optimal for linear functional without service interruption. If prevent the implementation of programs interrupted, sometimes absolute priority disciplines are better. However, in systems with many devices maintenance procedure can be disrupted, while retaining the properties of known subjects. For example, in case of a FIFO discipline, if a job is received after the service does not affect the incoming job before him, it may be served. This can be done using Batchfill mechanism (check available periods for low-priority tasks). Often it is required to provide manage the job not later or earlier, or in a specified period of time after entering the system. For a description of these conditions we consider the classification of the job queues on systems diagrams defining the service priorities.

Dr

peradze

jemal

Tbilisi state univer ...

participant

Doct ...

kalichava

zviadi

Tbilisi state University

presenter

Approximation with respect to the spatial variable of the solution of a nonlinear dynamic beam problem Z.Kalichava, J.Peradze We consider with the boundary value problem for a nonlinear integro-differential equation modeling the dynamic state of the Timoshenko beam. To approximate the solution with respect to the spatial variable, the Galerkin method is used, the error of which is estimated.

gwga ...

gwgausym

gwgausym

gwgausym

participant

72 people are registered