=Paper=
{{Paper
|id=Vol-2281/paper-14
|storemode=property
|title=A Web-based System for Launching Large Experiment Series on Supercomputers
|pdfUrl=https://ceur-ws.org/Vol-2281/paper-14.pdf
|volume=Vol-2281
|authors=Evgheniy Kuklin,Sergei Pravdin
}}
==A Web-based System for Launching Large Experiment Series on Supercomputers==
https://ceur-ws.org/Vol-2281/paper-14.pdf
A Web-based System for Launching
Large Experiment Series on Supercomputers?
Evgeniy Kuklin1,2 and Sergei Pravdin1,2
1
Krasovskii Institute of Mathematics and Mechanics, Ekaterinburg, Russia
key@imm.uran.ru
2
Ural Federal University, Ekaterinburg, Russia
Abstract. The researchers developed a user-friendly web-based service
for launching large series of experiments on parallel computing systems.
Simulation of various biological processes requires that dozens of nu-
merical experiments with parameter variations have to be conducted.
A key feature of the proposed system is the automatic generation of
configuration files based on a proposed algorithm for generating tuples
and automatic job launching, which saves researchers from manual data
preparation. The developed system is currently used for conducting com-
putational experiments to study the drift of spiral waves in myocardium
on the Uran supercomputer, Ekaterinburg, Russia.
Keywords: Parallel computing systems · Supercomputers · Experiment
series launching · Graphical user interface · Living system simulation
1 Introduction
Living system simulation often demands a large number of computational exper-
iments on the same model but with varying parameter values. Since experiments
are time-consuming, conducting them in a reasonable time frame without parallel
computing systems and supercomputers is difficult. In order to obtain accurate
results, dozens or hundreds of numerical experiments must be prepared and per-
formed. For many scientists, working with a command line on supercomputers
is tedious and frustrating.
Often, these problems are solved using already existing specialized applica-
tions, which are accessed from various web platforms. However, most platforms
are focused on a specific task and utilize preinstalled software, which limits their
use. In addition, they do not allow users to flexibly configure the automatic gen-
eration of input data for different simulation models. This can cause problems
when conducting a large number of experiments.
To reduce experiment preparation, the authors developed the web-based mul-
tiplatform system for launching experiments on a supercomputer. The system
conducts large series of computational experiments with different parameter val-
ues, saving scientists from manually preparing the input data and launching jobs.
?
Supported by the RSF project 17-71-20024 (IMM UB RAS).
System for Launching Large Experiment Series on Supercomputers 137
It tracks and saves parameters of previous experiments to be reused or reconfig-
ured. The service provides a simple web interface that works remotely without
the installation of additional software. Though the project was initiated for the
purposes of heart modeling, the system can also be used for different computa-
tional clusters and can run almost any software that has an input configuration
file in the standard INI file format. A build-in parser supports sections, enclosed
in square brackets and making parameter lists more structured, and comments.
If varied parameter values make an arithmetic progression, they can be specified
in a short form as a start value, an increment, and a final value. The present ar-
ticle describes the architecture of the developed system, as well as the algorithm
for generating tuples from sets of values of individual parameters, and features
of the graphical user interface.
2 Service Architecture
After we encountered technical problems when using Python for desktop GUI
development, we conceptualized the interface. Remote users could use different
operation systems, and could not have administrator rights, which caused prob-
lems with installing the required packages for the desktop version. Therefore, a
web-based interface was conceived, which should work with any computer, even
on a tablet or smartphone.
The schematic diagram of the system is shown in Fig. 1. The main com-
ponents of the system are located on a dedicated server. These include: a web
server, which provides the interface, a database, a storage directory for experi-
mental metadata, and several service scripts.
User PC Dedicated server
HTTP PHP
Web interface Web server Database
SFTP
Computational SLUR M
Head node
nodes
Supercomputer
NFSv4
/home Archive
Storage system
Fig. 1. Web service architecture
138 E. Kuklin and S. Pravdin
The system uses the concept of a computational package, which is prepared
via the interface and then sent through “SFTP” protocol to the supercomputer
storage system into the user’s home directory. The entire computational package
is stored in a special directory on the web server for possible re-use in the future;
key information about the package is written to the database. All supporting
scripts are sent along with the original data in the computational package, but
they can also be taken from a special directory on the storage system.
On the supercomputer, after the automatic preparation of input configura-
tion files for each experiment of the series (see Section 3.3), the launch script
sets the tasks for execution through interacting with the supercomputer work-
load manager, such as “SLURM”. Data obtained as a result of the calculations
remain in the storage system, which is connected to the compute nodes and the
cluster head node via “NFSv4” protocol. These data can optionally be placed in
a common archive of the research group, which work with the system.
3 Technical Issues
3.1 Web Interface
One of the main objectives of the project was to provide an easy-to-use system
through a user-friendly interface requiring minimal adaptation. As mentioned,
the universal web interface used likely on any computer.
The process of setting a series of experiments is divided into four steps. In
the first step (see Appendix; Fig. 2), a user can start a new project, select one
of the previous projects, or check the status of jobs already running on the
supercomputer.
The second step (Appendix; Fig. 3) involves editing the configuration file for
the simulation program. Simulation of living systems may involve the launch of a
large number of experiments on one model with a variation of several parameters.
For convenience, the parameter values for each experiment can be specified in
one file, separating the values with a space.
In the third step (Appendix; Fig. 4), the user specifies all the necessary
information to place the jobs in the execution queue on a parallel computing
system. The information includes the path to the source folder, parameters for
the workload manager, the working directory, and a description of the experi-
ment. Presumable, the source code is already stored on the supercomputer, since
the program must be built there.
The last step (Appendix; Fig. 5) displays details of the experiment series,
and the queue of the active user jobs (confirming the tasks have been success-
fully started for calculation). To display the queue, the supercomputer workload
manager receives a request with the appropriate command from the web server.
3.2 Database Usage
In order to repeat a previous experiment, it is sufficient to save the simulation
program, the experimental configuration files, and the parameters for the work-
load manager. Every experiment is assigned a unique identifier stored in the
System for Launching Large Experiment Series on Supercomputers 139
database along with other parameters of the experiment that becomes the name
of the directory storing the files related to the experiment. The rest of the data
is saved in the next steps of the experiment.
On the service home page, the latest experiments of all users are displayed
using information from the database (see Appendix; Fig. 2). This makes it pos-
sible to avoid duplicate calculations when required experiment has already been
conducted or to reproduce an experiments when necessary. Moreover, once per-
formed, a series of experiments can be easily restarted with different parameters
without replacing the fields, which accelerates launching repeated calculations.
The database used for the project is PostgreSQL [1]. The launching scripts
were written in Python, Bash, and PHP.
3.3 Generation of Configuration Files
The used simulation software stores all model parameters in a standard INI con-
figuration file. For example, to study the drift of spiral waves in the myocardium,
several models with 10–20 parameters are used, and almost every parameter in
the corresponding configuration files can be variated. The system never knows
how many combinations will result from user settings. Using a recursion is not a
good idea, so an efficient algorithm for full parameter enumeration was necessary.
Each input file describes values of parameters a0 , a1 , . . . , ak . According to
a specific task, a parameter ai can be constant for all launches or must take
values from a set Vi = {a0i , a1i , . . . , ani i }. So, a set of input files where varying
parameters take values from their respective sets must be created. We propose
a simple algorithm to get all tuples. The number of tuples is N = (n0 + 1)(n1 +
1) . . . (nk +1). Each input file is created based on indices i0 , i1 , . . . , ik of parameter
values. To get a tuple corresponding to an index I ∈ {0, . . . , N − 1}, we use a
backing array J1...k and utilize the algorithm 1.
Algorithm 1: Tuples generation algorithm
procedure FindTuples
Jk = I
for p from k − 1 downto 1 do
begin
ip = Jp+1 mod np
Jp = Jp+1 div np
end
i0 = J1 mod n0
Finally, the I th tuple is formed using indices i0 , i1 , . . . , ik , which serve to
create N configuration files with unique parameter values. Each configuration
file is placed to its own directory, from where its own copy of simulation software
is launched. For convenience, the directory names depend on values and names
of the varied parameters.
140 E. Kuklin and S. Pravdin
Two ways of specifying a set of values of one parameter are supported. Users
may simply list the values by separating them with a space or apply an expres-
sion, like 200...10...300, in which the initial value, the step, and the final value
are specified, respectively.
4 Discussion and Approbation
The users of the developed system appreciated the convenience of the new web
GUI and the ability to launch a series of dozens of experiments in five minutes
without working in a command line. Using settings from previous experiments
reduces the time to launch a new experiment series to less than a minute. Over-
all, the user-friendly system helped the researchers to conduct computational
experiments more efficiently.
The proposed system was tested with the Uran supercomputer located at
the Krasovskii Institute of Mathematics and Mechanics of the Russian Academy
of Sciences in Ekaterinburg, Russia. The system currently carries out computa-
tional experiments to study the drift of spiral waves in myocardium. Myocardium
is an active medium and consists of interconnected elements. They have a rest-
ing state and can temporarily reach an excited state when enough stimulus is
applied. Excited elements produce stimuli that spread in all directions and can
excite neighboring elements. Thus, waves of excitation can appear. The waves
can be plane and spiral. Spiral waves in myocardium emerge only in the case
of dangerous arrhythmias and must be treated if they persist. One treatment
called low-voltage cardioversion-defibrillation (LVCD) involves periodical apply-
ing a small electrical current to an area in the myocardium so that the stimulated
zone produces plane waves with greater frequency than the spiral waves. The
plane waves begin to occupy broader and broader zones and finally supersede
the spiral waves. The aim of the study is to find out the optimal stimulation
parameters for the LVCD. Five models with different numbers of parameters are
used for the simulation.
The induced drift of spiral waves was simulated with a range of cardiac
models, model parameters, stimulation periods, and areas [2]. There were 6·8+5·9
sets of parameters in total (parameters varied for different models).
As myocardium has fibers along which the excitation spreads faster than
across, it considered anisotropic. The wave drift was studied in an anisotropic
tissue [3], so fiber direction was a new parameter that also varied. In total, 7 · 5 · 2
parameter sets were examined.
Another work was devoted to studying LVCD in anisotropic myocardium
models with curved fibers using biophysical ionic Luo—Rudy cell model [4].
After measuring the time of the spiral wave drift and determining the type of
interaction between waves, the findings were compared with the results of the
isotropic and parallel fiber anisotropic cases. In total, seventeen parameter sets
were investigated.
The computational program was written in C using a third-party software
[5] for parsing INI files.
System for Launching Large Experiment Series on Supercomputers 141
Although the service was designed for heart modeling problems, it can also
be used with another computational clusters and other software.
5 Related Work
To assist domain-specific scientists in conducting numerical experiments on par-
allel computing systems, various platforms have been created [6]. A number
of projects have provided the integration of application software packages with
supercomputers. For example, the DiVTB (Distributed Virtual Test Bed) plat-
form provides a task-oriented approach for solving specific classes of problems
in computer-aided engineering through resources supplied by grid computing
environments. It has a user-friendly graphical interface where parameters of a
computational experiment can be specified, and the experiment can then be
executed on a supercomputer.
A “Specialized web portal for solving problems on multiprocessor computing
systems” [7] is a similar to DiVTB project for remote calculations. The system
incorporates several parallel algorithms to solve the inverse gravity of lateral
density reconstruction, the structural inverse gravity, and the magnetic problem
of the contact surface reconstruction.
A platform called Education-research Integration through Simulation On the
Net” (EDISON) [8] has been designed and implemented to access and run vari-
ous technological computer-aided design software tools. The platform provides an
easy-to-use GUI that helps geographically distributed researchers run and share
their tools in five areas: computational fluid dynamics, computational chemistry,
nanophysics, computer-aided optimal design, and computational structural dy-
namics.
The Orion system [9] provides a practical and economical interface on the
Tianhe-2 supercomputer to enable big data applications to run on Tianhe-2 via
a single command or a shell script.
While many systems are limited by an integrated set of algorithms or ap-
plications, the distinctive feature of the proposed system is the ability to use
various software to run experiments, providing users flexibility and convenience.
Also, previous projects did not provide automation, such as launching of a se-
ries of computational experiments with varying parameter values. However, the
positive experience of management of such systems via a web interface is worth
noting and inspired the creation of the proposed system.
6 Conclusion
The paper describes the web-based multiplatform system for launching a series
of experiments on a supercomputer. The service interface allows users to set the
parameters and run jobs on a supercomputer without working in a command
line. The system focuses on minimizing the number of user actions required for
launching a large series of experiments. Using the built-in database, a previously
conducted experiment can be quickly restarted with new settings. The system
142 E. Kuklin and S. Pravdin
supports INI sections and comments, and allows users to specify a range of values
incrementally.
The architecture of interaction among a computing cluster, a database, and
an external web application were presented with the algorithm for generating tu-
ples from sets of values with individual parameters, and features of the graphical
user interface were also presented. The system is used for carrying out computa-
tional experiments to study the drift of spiral waves in myocardium on the Uran
supercomputer.
The nearest extension of the project is the option of automatically writing
to the archive with indexing in the database and the ability to search through
previously launched experiments using keywords. Adapting the system for more
interactive job control (for example, canceling all the jobs launched during an
experiment) will greatly enhance usability. The researchers also plan to add
build-in data post-processing methods for convenience.
Acknowledgments
Our study was performed using the Uran supercomputer of the Krasovskii In-
stitute of Mathematics and Mechanics.
References
1. PostgreSQL: The world’s most advanced open source database.
https://www.postgresql.org/. Accessed: 2018-06-29.
2. Sergei Pravdin, Timur Nezlobinsky, and Alexander Panfilov. Inducing drift of spi-
ral waves in 2D isotropic model of myocardium by means of an external stimu-
lation. CEUR Workshop Proceedings, 1894:268 – 284, 2017. URL http://ceur-
ws.org/Vol-1894/bio6.pdf. Proceedings of the 48th International Youth School-
Conference ’Modern Problems in Mathematics and its Applications’, Yekaterinburg,
Russia, February 5-11, 2017.
3. Timofei Epanchintsev, Sergei Pravdin, Andrey Sozykin, and Alexander Panfilov.
Simulation of overdrive pacing in 2D phenomenological models of anisotropic my-
ocardium. Procedia Computer Science, 119:245 – 254, 2017. ISSN 1877-0509. URL
http://www.sciencedirect.com/science/article/pii/S187705091732392X. 6th
International Young Scientist Conference on Computational Science, YSC 2017,
01-03 November 2017, Kotka, Finland.
4. Timofei Epanchintsev, Sergei Pravdin, and Alexander Panfilov. Spiral wave drift
induced by high-frequency forcing. Parallel simulation in the Luo–Rudy anisotropic
model of cardiac tissue. In Computational Science – ICCS 2018, pages 378–391.
Springer International Publishing, 2018.
5. N. Devillard. IniParser: stand-alone ini parser library in ANSI C.
https://github.com/ndevilla/iniparser, 2016.
6. Stanislav P.Polyakov, Andrey P.Demichev, and Alexander P.Kryukov. Web toolkit
for scientific research: State of the art and the prospect for development. Procedia
Computer Science, 66:429–438, 2015.
System for Launching Large Experiment Series on Supercomputers 143
7. Elena Akimova, Vladimir Misilov, Aliya Skurydina, and Maxim Martyshko. Special-
ized web portal for solving problems on multiprocessor computing systems. CEUR
Workshop Proceedings, 1513:123 – 129, 2015. URL http://ceur-ws.org/Vol-1513/
paper-12.pdf. Proceedings of the 1st Ural Workshop on Parallel, Distributed, and
Cloud Computing for Young Scientists, Yekaterinburg, Russia, Nov. 17th, 2015.
8. Suh Young-Kyoon, Hoon Ryu, Hanki Kim, and Kum Won Cho. EDISON: A web-
based hpc simulation execution framework for large-scale scientific computing soft-
ware. pages 608 – 612, 2016. Proceeding of IEEE/ACM 16th International Sympo-
sium on Cluster Cloud and Grid Computing (CCGrid 2016), May 2016.
9. Xi Yang, Chengkun Wu, Kai Lu, Lin Fang, Yong Zhang, Shengkang Li, Guixin Guo,
and YunFei Du. An interface for biomedical big data processing on the Tianhe-2
supercomputer. Molecules, 22(12), 2017. URL http://www.mdpi.com/1420-3049/
22/12/2116/html.
Appendix
Typical screens from the web interface are listed below.
Fig. 2. GUI. Step 1
144 E. Kuklin and S. Pravdin
Fig. 3. GUI. Step 2 Fig. 4. GUI. Step 3
System for Launching Large Experiment Series on Supercomputers 145
Fig. 5. GUI. Step 4