Conference Program

MONDAY, August 21

Arrivals, private meetings

TUESDAY, August 22

08:00-09:00

Registration and Welcome Coffee

09:00-

09:05

OPENING SESSION 

room C1.1.334L

09:05-

11:15

Frameworks and Services                                   Chair: Iosif Meyerov

room C1.1.334L

09:05-09:30

Pair of Genes: Technical Validation of Distributed Causal Role Attribution to Gene Network Expansion
Diana Dolgaleva, Camilla Pelagalli, Stefania Pilati, Enrico Blanzieri, Valter Cavecchia, Sergey Astafiev, and Alexander Rumyantsev

(remote)

09:30-09:55

HiTViSc: High-Throughput Virtual Screening as a Service
Natalia Nikitina and Evgeny Ivashko

09:55-10:20

Expanding the Cellular Automata Topologies Library for Parallel Implementation of Synchronous Cellular Automata
Yuri Medvedev, Sergey Kireev, and Yulia Trubitsyna

10:20-10:45

Parallel-batched Interpolation Search Tree
Vitaly Aksenov, Ilya Kokorin, and Alena Martsenyuk

10:45-

11:10

COFFEE BREAK

11:10-

12:50

Hardware and System Software                       Chair: Natalia Nikitina

room C1.1.334L

11:10-

11:35

Current trends in the high performance computing

Alexander Moskovsky,  RSC Group

11:35-12:00

Trace Balancing Technique for Trace Playback in LuNA System
Victor Malyshkin, Vladislav Perepelkin, and Artem Lyamin

12:00-12:25

Case Study for Running Memory-Bound Kernels on RISC-V CPUs
Valentin Volokitin, Evgeny Kozinov, Valentina Kustikova, Alexey Liniov, and Iosif Meyerov

12:25-12:50

Automation of programming for promising high-performance computing systems
Vladimir Bakhtin, Dmitry Zakharov, Nikita Kataev, Alexander Kolganov, and Mikhail Yakobovskiy

12:50-

14:00

LUNCH 

AITU Cafeteria

14:00-

15:30

Distributed Systems Management                      Chair: Nikita Kataev

room C1.1.334L

14:00-14:25

Probabilistic Resources Allocation with Group Dependencies in Distributed Computing

Victor Toporkov, Dmitry Yemelyanov, and Artem Bulkhak

(remote)

14:25-14:50

Verifying the Correctness of HPC Performance Monitoring Data

Danil Kashin and Vadim Voevodin

14:50-15:15

Scheduling of Workflows with Task Resource Requirements in Cluster Environments

Oleg Sukhoroslov

(remote)

15:15-15:40

Multicriteria Task Distribution Problem for Resource-Saving Data Processing

Anna Klimenko and Arseniy Barinov

(remote)

15:40-

16:00

COFFEE BREAK

16:00-

17:30

Distributed Algorithms                                    Chair: Vadim Voevodin

room C1.1.334L

16:00-16:25

Parallel factorization methods in domain decomposition : Slides

Valery Il’in

(remote)

16:25-16:50

Automatic parallelization of iterative loops nests on distributed memory computing systems
Anton Bagliy, Elena Metelitsa, and Boris Steinberg

(remote)

16:50-17:15

Didal: Distributed Data Library for Development of Parallel Fragmented Programs
Victor Malyshkin and Georgy Schukin

(remote)

17:15-17:40

Combinatorial aspect of code restructuring for virtual memory computer systems under WS swapping strategy
Stepan Vyazigin and Madina Mansurova

(remote)

17:40-18:05

Parallel Generation and Analysis of Optimal Chordal Ring Networks Using Python Tools on Kunpeng Processors
Oleg Monakhov, Emilia Monakhova, and Sergey Kireev

19:00-

22:00

Conference Dinner

WEDNESDAY, August 23

08.00-20.00

SOCIAL DAY : 

Full day excursion: Korgalzhyn Nature Reserve

THURSDAY, August 24

Workshop Day

10:00-11:00

Workshop: Practical Introduction to LuNA System
Automatic construction of numerical parallel programs from a high-level portable specification in a simple way. A brief introduction and a series of practical exercises in a terminal class to construct and execute parallel programs using LuNA system.
Vladislav Perepelkin, ICM&MG SB RAS, Russia

room: C1.1.357K

11:00-

11:30

COFFEE BREAK

11:30-

13:00

Advisors and Young Scientists                Chair: Sergey Kireev

room: C1.2.233L

11:30-11:45

Advisor’s Intro
Computing device community forming in tasks of distributed data processing based on the resource consumption estimations
Anna Klimenko, Institute of IT and Security Technologies, RSUH, Russia

11:45-11:57

The Practice of Pareto frontier forming for the distributed computing system resources optimization
Arseny Barinov, Institute of IT and Security Technologies, RSUH, Russia

11:57-12:08

Advisor’s Intro
Overview of Currents Works on the LuNA Project
Vladislav Perepelkin, ICM&MG SB RAS, Novosibirsk State University, Russia

12:08-12:20

Efficient Fragmented Implementation of a Fluid Filtration Problem Solver Using MPI
Vitaly Spirin, ICM&MG SB RAS, Novosibirsk State University, Russia

12:20-12:35

Advisor’s Intro
Automated Parallelization and Optimization of Programs
Boris Ya. Steinberg, Southern Federal University, Institute of Mathematics, Mechanics and Computer Science, Russia

12:35-12:47

Automation Optimization of the Gauss-Seidel algorithm for solving the generalized Dirichlet problem
E. A. Metelitsa, Southern Federal University, Institute of Mathematics, Mechanics and Computer Science, Russia

12:47-13:00

Parameterized Automated Parallelization of a Branch-and-Bound Algorithm
V. V. Burkhovetskiy, Southern Federal University, Institute of Mathematics, Mechanics and Computer Science, Russia

13:00-

14:00

LUNCH

AITU Cafeteria

14:00-15:30

Roundtable: Trends in Parallel Computing Technologies
All participants present their views

Moderator: Maxim Gorodnichev, ICM&MG SB RAS, Russia

room: C1.2.233L

15:30-

16:00

COFFEE BREAK

16:00-17:00

Workshop: Specialized Independent Operating System for Supercomputing Centers: prospects and requirements 

OpenScaler distribution general overview
Dmitry Varenov, Huawei

Key OpenScaler innovative software
Vladimir Pavlov, Huawei

OpenScaler OS for Science and Education: Experience
Maxim Gorodnichev, ICM&MG SB RAS, Russia

Talks and Discussion: Prospects and Requirements for Specialized OS in Supercomputing
All participants

room: C1.2.233L

17:00-19:00

Workshop: Practical Introduction to DVM System
The DVMH parallel programming model provides high-level specifications to expose parallelism in C and Fortran code. The advanced runtime system manages the program execution and adapts it to all available resources. Since it is aware of program execution as well as higher-level parallelism specifications, the DVM system provides tools for program debugging and performance analysis. We will give a brief introduction to the DVMH programming model and help participants to go through the main steps of parallel program development in a practical way. We will also discuss the System FOR Automated Parallelization (SAPFOR) which is a software development suite that aims to produce a parallel version of a sequential program according to the DVMH programming model
Nikita Kataev, Alexander Kolganov, Keldysh Institute of Applied Mathematics RAS, Russia

room: C1.1.357K

FRIDAY, August 25

Private meetings, departures