ICALEPCS2023 - List of Authors (Forsberg, J.)

Paper	Title	Page
TUPDP044	Improving Performance of Taranta: Analysis of Memory Requests and Implementation of the Solution	617
	M. Canzari INAF - OAAB, Teramo, Italy V. Alberti INAF-OAT, Trieste, Italy A. Dubey PSL, Pune, India M. Eguiraun, J. Forsberg, V. Hardion MAX IV Laboratory, Lund University, Lund, Sweden A. Georgiou CGI, Edinburgh, United Kingdom H.R. Ribeiro Universidade do Porto, Faculdade de Ciências, Porto, Portugal
	Taranta is a software suite for generating graphical interfaces for Tango Controls software, currently adopted by MaxIV for scientific experiment usage, SKA during the current construction phase for the development of engineering interfaces for device debugging, and other institutions. A key feature of Taranta is the ability to create customizable dashboards without writing code, making it easy to create and share views among users by linking the dashboards to their own tango devices. However, due to the simplicity and capabilities of Taranta’s widgets, more and more users are creating complex dashboards, which can cause client-side resource problems. Through an analysis of dashboards, we have found that excessive memory requests are generated by a large amount of data. In this article, we report on the process we believe will help us solve this performance issue. Starting with an analysis of the existing architecture, the issues encountered, and performance tests, we identify the causes of these problems. We then study a new architecture exploiting all the potential of the Javascript framework React on which Taranta is built, before moving on to implementation of the solution.
	Poster TUPDP044 [1.549 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-TUPDP044
About •	Received ※ 04 October 2023 — Revised ※ 18 October 2023 — Accepted ※ 14 December 2023 — Issued ※ 16 December 2023
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

THMBCMO01	New Developements on HDB++, the High-performance Data Archiving for Tango Controls	1190
	D. Lacoste, R. Bourtembourg ESRF, Grenoble, France J. Forsberg MAX IV Laboratory, Lund University, Lund, Sweden T. Juerges SKAO, Macclesfield, United Kingdom J.J.D. Mol ASTRON, Dwingeloo, The Netherlands L. Pivetta, G. Scalamera Elettra-Sincrotrone Trieste S.C.p.A., Basovizza, Italy S. Rubio-Manrique ALBA-CELLS, Cerdanyola del Vallès, Spain
	The Tango HDB++ project is a high performance event-driven archiving system which stores data with micro-second resolution timestamps. HDB++ supports many different backends, including MySQL/MariaDB, TimeScaleDB (a time-series PostgreSQL extension), and soon SQLite. Building on its flexible design, latest developments made supporting new backends even easier. HDB++ keeps improving with new features such as batch insertion and by becoming easier to install or setup in a testing environment, using ready to use docker images and striving to simplify all the steps of deployment. The HDB++ project is not only a data storage installation, but a full ecosystem to manage data, query it, and get the information needed. In this effort a lot of tools were developed to put a powerful backend to its proper use and be able to get the best out of the stored data. In this paper we will present as well the latest developments in data extraction, from low level libraries to web viewer integration such as grafana. Pointing out strategies in use in terms of data decimation, compression and others to help deliver data as fast as possible.
	Slides THMBCMO01 [0.926 MB]
	Poster THMBCMO01 [0.726 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-THMBCMO01
About •	Received ※ 05 October 2023 — Revised ※ 24 October 2023 — Accepted ※ 08 December 2023 — Issued ※ 16 December 2023
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

THPDP050	Improving User Experience and Performance in Sardana and Taurus: A Status Report and Roadmap	1420
	Z. Reszela, J. Aguilar Larruy, M. Caixal i Joaniquet, G. Cuní, R. Homs-Puron, E. Morales, M. Navarro, C. Pascual-Izarra, J.A. Ramos, S. Rubio-Manrique, O. Vallcorba ALBA-CELLS, Cerdanyola del Vallès, Spain B. Bertrand, J. Forsberg MAX IV Laboratory, Lund University, Lund, Sweden M.T. Núñez Pardo de Vera DESY, Hamburg, Germany M. Piekarski NSRC SOLARIS, Kraków, Poland D. Schick MBI, Berlin, Germany
	Sardana Suite is an open-source scientific SCADA solution used in synchrotron light beamlines at ALBA, DESY, MAXIV and SOLARIS and in laser labs at MBI-Berlin. It is formed by Sardana and Taurus - both mature projects, driven by a community of users and developers for more than 10 years. Sardana provides a low level interface to the hardware, middle level abstractions and a sequence engine. Taurus is a library for developing graphical user interfaces. Sardana Suite uses client - server architecture and is built on top of TANGO. As a community, during the last few years, on one hand we were focusing on improving user experience, especially in terms of reliability and performance and on the other hand renewing the dependency stack. The system is now more stable, easier to debug and recover from a failure. An important effort was put in profiling and improving performance of Taurus applications startup. The codebase has been migrated to Python 3 and the plotting widgets were rewritten with pyqtgraph. This didn’t prevent us from delivering new features, like for example the long-awaited configuration tools and format based on YAML which is easy and intuitive to edit, browse, and track historical changes. Now we conclude this phase in the project’s lifetimes and are preparing for new challenging requirements in the area of continuous scans like higher data throughput and more complex synchronization configurations. Here we present the status report and the future roadmap.
	Poster THPDP050 [0.605 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-THPDP050
About •	Received ※ 06 October 2023 — Revised ※ 26 October 2023 — Accepted ※ 13 December 2023 — Issued ※ 21 December 2023
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

FR2BCO03	Taranta Project - Update and Current Status	1657
	Y.L. Li, M. Eguiraun, J. Forsberg, V. Hardion, M. Leorato MAX IV Laboratory, Lund University, Lund, Sweden V. Alberti INAF-OAT, Trieste, Italy M. Canzari INAF - OAAB, Teramo, Italy A. Dubey PSL, Pune, India M. Gandor, D.T. Trojanowska S2Innovation, Kraków, Poland H.R. Ribeiro Universidade do Porto, Faculdade de Ciências, Porto, Portugal
	Taranta, developed jointly by MAX IV Laboratory and SKA Observatory, is a web based no-code interface for remote control of instruments at accelerators and other scientific facilities. It has seen a great success in system development and scientific experiment usage. In the past two years, the panel of users has greatly expanded. The first generation of Taranta was not able to handle the challenges introduced by the user cases, notably the decreased performance when a high number of data points are requested, as well as new functionality requests. Therefore, a series of refactoring and performance improvements of Taranta are ongoing, to prepare it for handling large data transmission between Taranta and multiple sources of information, and to provide more possibilities for users to develop their own dashboards. This article presents the status of the Taranta project from the aspects of widgets updates, packages management, optimization of the communication with the backend TangoGQL, as well as the investigation on a new python library compatible with the newest python version for TangoGQL. In addition to the technical improvements, more facilities other than MAX IV and SKAO are considering to join Taranta project. One workshop has been successfully held and there will be more in the future. This article also presents the lesson learned from this project, the road map, and the GUI strategy for the near future.
	Slides FR2BCO03 [4.759 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-FR2BCO03
About •	Received ※ 06 October 2023 — Accepted ※ 21 November 2023 — Issued ※ 23 November 2023
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

Paper

Title

Page

Improving Performance of Taranta: Analysis of Memory Requests and Implementation of the Solution

617

M. Canzari
INAF - OAAB, Teramo, Italy
V. Alberti
INAF-OAT, Trieste, Italy
A. Dubey
PSL, Pune, India
M. Eguiraun, J. Forsberg, V. Hardion
MAX IV Laboratory, Lund University, Lund, Sweden
A. Georgiou
CGI, Edinburgh, United Kingdom
H.R. Ribeiro
Universidade do Porto, Faculdade de Ciências, Porto, Portugal

Taranta is a software suite for generating graphical interfaces for Tango Controls software, currently adopted by MaxIV for scientific experiment usage, SKA during the current construction phase for the development of engineering interfaces for device debugging, and other institutions. A key feature of Taranta is the ability to create customizable dashboards without writing code, making it easy to create and share views among users by linking the dashboards to their own tango devices. However, due to the simplicity and capabilities of Taranta’s widgets, more and more users are creating complex dashboards, which can cause client-side resource problems. Through an analysis of dashboards, we have found that excessive memory requests are generated by a large amount of data. In this article, we report on the process we believe will help us solve this performance issue. Starting with an analysis of the existing architecture, the issues encountered, and performance tests, we identify the causes of these problems. We then study a new architecture exploiting all the potential of the Javascript framework React on which Taranta is built, before moving on to implementation of the solution.

Poster TUPDP044 [1.549 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-TUPDP044

About •

Received ※ 04 October 2023 — Revised ※ 18 October 2023 — Accepted ※ 14 December 2023 — Issued ※ 16 December 2023

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

THMBCMO01

New Developements on HDB++, the High-performance Data Archiving for Tango Controls

1190

D. Lacoste, R. Bourtembourg
ESRF, Grenoble, France
J. Forsberg
MAX IV Laboratory, Lund University, Lund, Sweden
T. Juerges
SKAO, Macclesfield, United Kingdom
J.J.D. Mol
ASTRON, Dwingeloo, The Netherlands
L. Pivetta, G. Scalamera
Elettra-Sincrotrone Trieste S.C.p.A., Basovizza, Italy
S. Rubio-Manrique
ALBA-CELLS, Cerdanyola del Vallès, Spain

The Tango HDB++ project is a high performance event-driven archiving system which stores data with micro-second resolution timestamps. HDB++ supports many different backends, including MySQL/MariaDB, TimeScaleDB (a time-series PostgreSQL extension), and soon SQLite. Building on its flexible design, latest developments made supporting new backends even easier. HDB++ keeps improving with new features such as batch insertion and by becoming easier to install or setup in a testing environment, using ready to use docker images and striving to simplify all the steps of deployment. The HDB++ project is not only a data storage installation, but a full ecosystem to manage data, query it, and get the information needed. In this effort a lot of tools were developed to put a powerful backend to its proper use and be able to get the best out of the stored data. In this paper we will present as well the latest developments in data extraction, from low level libraries to web viewer integration such as grafana. Pointing out strategies in use in terms of data decimation, compression and others to help deliver data as fast as possible.

Slides THMBCMO01 [0.926 MB]

Poster THMBCMO01 [0.726 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-THMBCMO01

About •

Received ※ 05 October 2023 — Revised ※ 24 October 2023 — Accepted ※ 08 December 2023 — Issued ※ 16 December 2023

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

THPDP050

Improving User Experience and Performance in Sardana and Taurus: A Status Report and Roadmap

1420

Z. Reszela, J. Aguilar Larruy, M. Caixal i Joaniquet, G. Cuní, R. Homs-Puron, E. Morales, M. Navarro, C. Pascual-Izarra, J.A. Ramos, S. Rubio-Manrique, O. Vallcorba
ALBA-CELLS, Cerdanyola del Vallès, Spain
B. Bertrand, J. Forsberg
MAX IV Laboratory, Lund University, Lund, Sweden
M.T. Núñez Pardo de Vera
DESY, Hamburg, Germany
M. Piekarski
NSRC SOLARIS, Kraków, Poland
D. Schick
MBI, Berlin, Germany

Sardana Suite is an open-source scientific SCADA solution used in synchrotron light beamlines at ALBA, DESY, MAXIV and SOLARIS and in laser labs at MBI-Berlin. It is formed by Sardana and Taurus - both mature projects, driven by a community of users and developers for more than 10 years. Sardana provides a low level interface to the hardware, middle level abstractions and a sequence engine. Taurus is a library for developing graphical user interfaces. Sardana Suite uses client - server architecture and is built on top of TANGO. As a community, during the last few years, on one hand we were focusing on improving user experience, especially in terms of reliability and performance and on the other hand renewing the dependency stack. The system is now more stable, easier to debug and recover from a failure. An important effort was put in profiling and improving performance of Taurus applications startup. The codebase has been migrated to Python 3 and the plotting widgets were rewritten with pyqtgraph. This didn’t prevent us from delivering new features, like for example the long-awaited configuration tools and format based on YAML which is easy and intuitive to edit, browse, and track historical changes. Now we conclude this phase in the project’s lifetimes and are preparing for new challenging requirements in the area of continuous scans like higher data throughput and more complex synchronization configurations. Here we present the status report and the future roadmap.

Poster THPDP050 [0.605 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-THPDP050

About •

Received ※ 06 October 2023 — Revised ※ 26 October 2023 — Accepted ※ 13 December 2023 — Issued ※ 21 December 2023

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

FR2BCO03

Taranta Project - Update and Current Status

1657

Y.L. Li, M. Eguiraun, J. Forsberg, V. Hardion, M. Leorato
MAX IV Laboratory, Lund University, Lund, Sweden
V. Alberti
INAF-OAT, Trieste, Italy
M. Canzari
INAF - OAAB, Teramo, Italy
A. Dubey
PSL, Pune, India
M. Gandor, D.T. Trojanowska
S2Innovation, Kraków, Poland
H.R. Ribeiro
Universidade do Porto, Faculdade de Ciências, Porto, Portugal

Taranta, developed jointly by MAX IV Laboratory and SKA Observatory, is a web based no-code interface for remote control of instruments at accelerators and other scientific facilities. It has seen a great success in system development and scientific experiment usage. In the past two years, the panel of users has greatly expanded. The first generation of Taranta was not able to handle the challenges introduced by the user cases, notably the decreased performance when a high number of data points are requested, as well as new functionality requests. Therefore, a series of refactoring and performance improvements of Taranta are ongoing, to prepare it for handling large data transmission between Taranta and multiple sources of information, and to provide more possibilities for users to develop their own dashboards. This article presents the status of the Taranta project from the aspects of widgets updates, packages management, optimization of the communication with the backend TangoGQL, as well as the investigation on a new python library compatible with the newest python version for TangoGQL. In addition to the technical improvements, more facilities other than MAX IV and SKAO are considering to join Taranta project. One workshop has been successfully held and there will be more in the future. This article also presents the lesson learned from this project, the road map, and the GUI strategy for the near future.

Slides FR2BCO03 [4.759 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-ICALEPCS2023-FR2BCO03

About •

Received ※ 06 October 2023 — Accepted ※ 21 November 2023 — Issued ※ 23 November 2023

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)