Sergei Devadze
Research Subjects
Sergei Devadze has a total of 65 co-authors. Here are the top co-authors by number of shared publications:
- Artur Jutman
Tallinn University of Technology
33 shared publications - Raimund Ubar
Tallinn University of Technology
27 shared publications - Jaan Raik
Department of Computer Systems, Tallinn University of Technology, Tallinn, Estonia
16 shared publications - Konstantin Shibin
Testonica Lab, Tallinn, Estonia
10 shared publications - Igor Aleksejev
Tallinn Univ. of Technology, Tallinn, Estonia
9 shared publications - Anton Tsertov
Testonica Lab, Tallinn, Estonia
6 shared publications
Latest Publications
- 29 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 33 (11) , 2998-3007
Architectural Exploration and Implementation of CERN LHC Trigger Algorithm With FPGA
The upcoming high-luminosity upgrade of the Large Hadron Collider (LHC) at CERN will increase data rates to values far exceeding the capabilities of software-based processing systems. As a result, new methods are required to efficiently extract scientifically valuable information from the massive data streams produced by the LHC’s particle detectors. This article discusses the design of the tau lepton trigger algorithm using the field-programmable gate array (FPGA) technology. Given its complexity and demanding technical requirements, realizing the algorithm in FPGA is a challenging task. This article details the algorithm development using high-level synthesis (HLS), a technique to generate hardware descriptions from the C++ code. We discuss architectural solutions and optimizations explored during the design process, including algorithm partitioning and pipelining, optimization of pipeline stages, floorplanning, and probing of implementation strategies. The performed design space exploration helped to improve latency, solve area-related issues, and reduce routing congestion, enabling implementation of tau lepton trigger on FPGA.
- 29 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Very Large Scale Integration (VLSI) Systems
- Vol. 33 (11) , 2998-3007
Architectural Exploration and Implementation of CERN LHC Trigger Algorithm With FPGA
The upcoming high-luminosity upgrade of the Large Hadron Collider (LHC) at CERN will increase data rates to values far exceeding the capabilities of software-based processing systems. As a result, new methods are required to efficiently extract scientifically valuable information from the massive data streams produced by the LHC’s particle detectors. This article discusses the design of the tau lepton trigger algorithm using the field-programmable gate array (FPGA) technology. Given its complexity and demanding technical requirements, realizing the algorithm in FPGA is a challenging task. This article details the algorithm development using high-level synthesis (HLS), a technique to generate hardware descriptions from the C++ code. We discuss architectural solutions and optimizations explored during the design process, including algorithm partitioning and pipelining, optimization of pipeline stages, floorplanning, and probing of implementation strategies. The performed design space exploration helped to improve latency, solve area-related issues, and reduce routing congestion, enabling implementation of tau lepton trigger on FPGA.
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
Ways for board and system test to benefit from FPGA embedded instrumentation
With continually growing adoption of FPGA based designs, and more features and capabilities available in FPGAs, board and system level test applications can - and should - take advantage of FPGA embedded instrumentation. Such FPGA assisted tests not only improve manufacturing test but also allow extensive testing in the field to address issues such as no-fault-found syndrome by enabling the detection of environment-induced defects or latent defects (related to bad solder joint quality, aging, etc.). In this paper we are briefly reviewing different types of FPGA embedded instruments before specifically addressing synthetic instrumentation and focusing on example use cases. We will be discussing design for test considerations, quality of test and diagnostics, savings in test time and test cost, and levels of automation previously not associated with functional and performance / stress test development.
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
Ways for board and system test to benefit from FPGA embedded instrumentation
With continually growing adoption of FPGA based designs, and more features and capabilities available in FPGAs, board and system level test applications can - and should - take advantage of FPGA embedded instrumentation. Such FPGA assisted tests not only improve manufacturing test but also allow extensive testing in the field to address issues such as no-fault-found syndrome by enabling the detection of environment-induced defects or latent defects (related to bad solder joint quality, aging, etc.). In this paper we are briefly reviewing different types of FPGA embedded instruments before specifically addressing synthetic instrumentation and focusing on example use cases. We will be discussing design for test considerations, quality of test and diagnostics, savings in test time and test cost, and levels of automation previously not associated with functional and performance / stress test development.
- 1 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
IEEE 1687 Compliant Ecosystem for Embedded Instrumentation Access and In-Field Health Monitoring
The ongoing research in the adjoin fields of failure resilience, health monitoring, fault management has also shown a widening interest in IEEE 1687 Std. applications. In this paper we unveil IEEE 1687 compliant ecosystem that covers almost every aspect related to IJTAG compliant RSNs. With the help of ecosystem tools we present an extension to the standard network reconfiguration modules to illustrate the capabilities of asynchronous error detection schemes in IJTAG networks. In addition, in this paper as an example application of the Ecosystems' SW support we present an automatically generated tests for the suite of IEEE 1687 benchmark networks in Procedural Description Language (PDL).
- 1 September 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
IEEE 1687 Compliant Ecosystem for Embedded Instrumentation Access and In-Field Health Monitoring
The ongoing research in the adjoin fields of failure resilience, health monitoring, fault management has also shown a widening interest in IEEE 1687 Std. applications. In this paper we unveil IEEE 1687 compliant ecosystem that covers almost every aspect related to IJTAG compliant RSNs. With the help of ecosystem tools we present an extension to the standard network reconfiguration modules to illustrate the capabilities of asynchronous error detection schemes in IJTAG networks. In addition, in this paper as an example application of the Ecosystems' SW support we present an automatically generated tests for the suite of IEEE 1687 benchmark networks in Procedural Description Language (PDL).
- 1 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Hierarchical Timing-Critical Paths Analysis in Sequential Circuits
The conventional design techniques struggle with integration density and constantly strengthening requirements of today's nanometer technology. Timing-critical paths analysis is one of such tasks. It has applications in critical path identification, path delay fault simulation, circuit reliability analysis e.g. Bias Temperature Instability (BTI) induced aging, and in several others. In this paper, we propose a scalable simulation based hierarchical technique for explicit identification of true timing-critical paths in sequential circuits. We explore the circuits at two levels - at the flat gate-level and at a higher level as a network of modules or sub-circuits. The result of timing analysis carried out at the gate-level is used for calculating the delays on the topological critical paths through the network of higher level modules. To speed-up the module level timing analysis, the theory of Structurally Synthesized BDDs (SSBDD) is used. Experimental results demonstrate considerable speed-up of the SSBDD based timing analysis, compared to the flat gate-level analysis.
- 1 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
Hierarchical Timing-Critical Paths Analysis in Sequential Circuits
The conventional design techniques struggle with integration density and constantly strengthening requirements of today's nanometer technology. Timing-critical paths analysis is one of such tasks. It has applications in critical path identification, path delay fault simulation, circuit reliability analysis e.g. Bias Temperature Instability (BTI) induced aging, and in several others. In this paper, we propose a scalable simulation based hierarchical technique for explicit identification of true timing-critical paths in sequential circuits. We explore the circuits at two levels - at the flat gate-level and at a higher level as a network of modules or sub-circuits. The result of timing analysis carried out at the gate-level is used for calculating the delays on the topological critical paths through the network of higher level modules. To speed-up the module level timing analysis, the theory of Structurally Synthesized BDDs (SSBDD) is used. Experimental results demonstrate considerable speed-up of the SSBDD based timing analysis, compared to the flat gate-level analysis.
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 305-310
Parallel Critical Path Tracing Fault Simulation in Sequential Circuits
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 305-310
Parallel Critical Path Tracing Fault Simulation in Sequential Circuits
- Open Access
- 11 November 2017
- Published by Springer Nature
- p. 15-38
Designing Reliable Cyber-Physical Systems
- 11 November 2017
- Published by Springer Nature
- p. 15-38
Open AccessDesigning Reliable Cyber-Physical Systems
- 1 October 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
Marginal PCB assembly defect detection on DDR3/4 memory bus
A contemporary high-performance system board is a complex 3D object that may contain dozens of hidden layers, stacked microvias, high density interconnect, with all of the above not contributing to the ease of test and reliability. High-speed signals are normally fine-tuned or even calibrated to deliver pitch perfect timing even in the case of now-ubiquitous DDR3 memories. Today, data transmission rates on the board may be reaching multigigabit ranges on a single channel. Such defects like dewetting, cold solder, head-in-pillow, voiding/crack in micro-via or excessive solder voids may result in system performance issues, increased error rates, intermittent faults and other sporadic stability issues observed in certain operation modes, at certain workloads or manifesting in a seemingly stochastic manner. In this work, we present a methodology aiming at systematic discovery of marginal defects, timing related faults and stability issues in board assembly's DDR3/4 data bus as part of the end-of-line testing in volume production environment. The methodology presented in this paper is based on an extended DDR memory controller in which we have converted the bus calibration mechanism into a fine-grain diagnostic instrument, which instead of masking marginal discrepancies (the normal case) would unveil and report them.
- 1 October 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
Marginal PCB assembly defect detection on DDR3/4 memory bus
A contemporary high-performance system board is a complex 3D object that may contain dozens of hidden layers, stacked microvias, high density interconnect, with all of the above not contributing to the ease of test and reliability. High-speed signals are normally fine-tuned or even calibrated to deliver pitch perfect timing even in the case of now-ubiquitous DDR3 memories. Today, data transmission rates on the board may be reaching multigigabit ranges on a single channel. Such defects like dewetting, cold solder, head-in-pillow, voiding/crack in micro-via or excessive solder voids may result in system performance issues, increased error rates, intermittent faults and other sporadic stability issues observed in certain operation modes, at certain workloads or manifesting in a seemingly stochastic manner. In this work, we present a methodology aiming at systematic discovery of marginal defects, timing related faults and stability issues in board assembly's DDR3/4 data bus as part of the end-of-line testing in volume production environment. The methodology presented in this paper is based on an extended DDR memory controller in which we have converted the bus calibration mechanism into a fine-grain diagnostic instrument, which instead of masking marginal discrepancies (the normal case) would unveil and report them.
- 11 September 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Design & Test
- Vol. 34 (6) , 27-35
Health Management for Self-Aware SoCs Based on IEEE 1687 Infrastructure
Motivated by the need to tolerate faults, this paper presents a complete fault management solution that includes fault detection and categorization, maintaining a map of faults, and modified scheduling and application algorithms for using healthy resources only. As the system maintains fairly sophisticated models of itself regarding faulty and healthy resources, it constitutes a good example of specialized self-awareness. —Axel Jantsch, TU Wien
- 11 September 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Design & Test
- Vol. 34 (6) , 27-35
Health Management for Self-Aware SoCs Based on IEEE 1687 Infrastructure
Motivated by the need to tolerate faults, this paper presents a complete fault management solution that includes fault detection and categorization, maintaining a map of faults, and modified scheduling and application algorithms for using healthy resources only. As the system maintains fairly sophisticated models of itself regarding faulty and healthy resources, it constitutes a good example of specialized self-awareness. —Axel Jantsch, TU Wien
- 1 September 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
Embedded instrumentation toolbox for screening marginal defects and outliers for production
High-performance communication interfaces in modern complex electronic systems bring new challenges for board-level manufacturing test. Various means of calibration and error-correction that are incorporated into high-speed protocols tend to hide manufacturing flaws thus making comprehensive testing be an extremely hard task. Links transmitting at gigabit rates are especially susceptible to so-called marginal defects. Marginal defect often escape detection by commonly used test methods as they manifest themselves only under certain conditions that may sporadically occur during device operation in the field. As a result, undetected marginal defects badly affect system stability and reduce its overall operating margin. In this paper, we present a novel embedded instrumentation toolbox intended for use in volume production environments to screen out outliers and marginal devices. The paper focuses on instruments for testing marginal defects on High-Speed Serial Links and DDR3/4 interfaces.
- 1 September 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
Embedded instrumentation toolbox for screening marginal defects and outliers for production
High-performance communication interfaces in modern complex electronic systems bring new challenges for board-level manufacturing test. Various means of calibration and error-correction that are incorporated into high-speed protocols tend to hide manufacturing flaws thus making comprehensive testing be an extremely hard task. Links transmitting at gigabit rates are especially susceptible to so-called marginal defects. Marginal defect often escape detection by commonly used test methods as they manifest themselves only under certain conditions that may sporadically occur during device operation in the field. As a result, undetected marginal defects badly affect system stability and reduce its overall operating margin. In this paper, we present a novel embedded instrumentation toolbox intended for use in volume production environments to screen out outliers and marginal devices. The paper focuses on instruments for testing marginal defects on High-Speed Serial Links and DDR3/4 interfaces.
- 1 November 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
A suite of IEEE 1687 benchmark networks
The saturation of the IJTAG concept and its approval as the IEEE 1687 standard in 2014 has generated a wave of research activities and created demand for a set of appropriate and challenging benchmarks. This paper presents such a set developed by an industrial and academic consortium and constructed in a way that facilitates objective comparison of experimental results across research groups as well as represents challenging network examples exhaustively utilizing features and constructs defined by the standard. The suite is arranged in four comprehensive categories, each having its particular purpose and composition principles, as described in the paper. We have also made an analysis of limitations of previous popular and ad-hoc benchmark sets as these limitations majorly motivated our current action. The new public-domain benchmarks are distributed together with source files and documentation through the dedicated web site. Some of the previous research results on IEEE 1687 have been reapplied on the new benchmarks set, thus creating an important initial reference point for the research community.
- 1 November 2016
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
A suite of IEEE 1687 benchmark networks
The saturation of the IJTAG concept and its approval as the IEEE 1687 standard in 2014 has generated a wave of research activities and created demand for a set of appropriate and challenging benchmarks. This paper presents such a set developed by an industrial and academic consortium and constructed in a way that facilitates objective comparison of experimental results across research groups as well as represents challenging network examples exhaustively utilizing features and constructs defined by the standard. The suite is arranged in four comprehensive categories, each having its particular purpose and composition principles, as described in the paper. We have also made an analysis of limitations of previous popular and ad-hoc benchmark sets as these limitations majorly motivated our current action. The new public-domain benchmarks are distributed together with source files and documentation through the dedicated web site. Some of the previous research results on IEEE 1687 have been reapplied on the new benchmarks set, thus creating an important initial reference point for the research community.