Jorn W Janneck

Followers

Following

Public Views

Interests

Uploads

Papers by Jorn W Janneck

Translating dataflow programs to efficient hardware: an MPEG-4 Simple Profile decoder case study

HAL (Le Centre pour la Communication Scientifique Directe), Mar 10, 2008

VL D dec ode engin e. int( size= VLD_ TABLE _ADDR _BITS ) vl d_in dex; int( size= VLD_ TABLE _DAT... more VL D dec ode engin e. int( size= VLD_ TABLE _ADDR _BITS ) vl d_in dex; int( size= VLD_ TABLE _DATA _BITS ) vl d_co dewor d := 1; proce dure star t_vld _engi ne( i nt in dex ) begin vld _inde x := inde x; vld _code word := 2 ; end funct ion v ld_s ucces s() --> bo ol : bita nd(vl d_cod eword ,3) = 0 e nd funct ion v ld_f ailur e() --> bo ol : bita nd(vl d_cod eword ,1) = 1 e nd funct ion v ld_r esult () --> int ( siz e=VL D_TAB LE_DA TA_BI TS ) : rs hift( vld_c odewo rd,2) end actio n bit s:[ b ] = => guard bit and(v ld_c odewo rd,3) = 2 do vld _code word := v ld_ta ble[ vld_i ndex + if b th en 1 else 0 en d ]; vld _inde x := rshi ft(vl d_cod eword ,2); bit _coun t := bit_ count + 1; end vld_f ailur e: a ction ==> guard vld _fail ure( ) // do // p rintl n("B ad VL D cod eword "); end gener ic_do ne: actio n ==> guard don e_rea ding _bits () end gener ic_do ne_w ith_b itrea d: ac tion bits :[b] ==> guard don e_rea ding _bits () do bit _coun t := bit_ count + 1; end // st uck: acti on == > // do // set_b its_ to_re ad(0) ; // reset _tex ture( ); // print ln(" Stuck at b it_co unt " + b it_co unt); // en d test_ zero_ byte : act ion = => guard don e_rea ding _bits (), bit and( read _resu lt(), 255 ) = 0 end test_ one_b yte: acti on == > guard don e_rea ding _bits (), bit and( read _resu lt(), 255 ) = 1 end reque st_by te: actio n ==> guard don e_rea ding _bits () do set _bits _to_ read( 8 ); end reque st_vo l: a ction ==> do set _bits _to_ read( VOL_ START _CODE _LEN GTH ) ; end sched ule f sm s tuck_ 1a /* look _for_ vo * / : // Proce ss a new VOL loo k_for _vo ( l ook_f or_vo ) --> vo_he ader; vo_ heade r ( g eneri c_don e ) --> stuck ; vo_ heade r ( v o_hea der.g ood ) --> look_ for_v ol; // Proce ss a new VOL loo k_for _vol ( g eneri c_don e ) --> stuck ; loo k_for _vol ( v ol_st artco de.g ood ) --> vol_o bject ; vol _obje ct ( v ol_ob ject_ laye r_ide ntifi catio n ) --> vol_a spect ; vol _aspe ct ( v ol_as pect. deta iled ) --> vol_c ontro l; vol _aspe ct ( g eneri c_don e ) --> vol_c ontro l; vol _cont rol ( v ol_co ntrol .det ailed ) --> vol_v bv; vol _cont rol ( g eneri c_don e_wi th_bi tread ) --> vol_s hape; vol _vbv ( v ol_vb v.det aile d ) --> vol_s hape; vol _vbv ( g eneri c_don e_wi th_bi tread ) --> vol_s hape; vol _shap e ( v ol_sh ape ) --> vol_t ime_i nc_re s; vol _time _inc _res ( v ol_ti me_in c_re s ) --> vol_w idth; vol _widt h ( v ol_wi dth ) --> vol_h eight ; vol _heig ht ( v ol_he ight ) --> vol_m isc; // vol_m isc ( vol _misc .uns uppor ted ) --> st uck; vol _misc ( g eneri c_don e ) --> look_ for_v op; // Proce ss a new VOP loo k_for _vop

Download

Mobies Position Paper (DRAFT 2 Version)

This document is responding to the MoBIES "Automotive Challenge Problems". It is prepared by the ... more This document is responding to the MoBIES "Automotive Challenge Problems". It is prepared by the Mobies Phase 1 Berkeley team, whose project is entitled "Process-Based Software Components for Networked Embedded Systems." The problems posed in the "Automotive Challenge Problems" paper are addresses one by one and our views on the problems are presented. As part of the Mobies phase I effort at Berkeley, we are developing a software framework called Ptolemy II. Ptolemy II is a component-oriented modeling and design framework written in Java. It is intended primarily to facilitate experimentation with design techniques and methodologies. We believe that a number of the challenge problems posed by the Mobies phase I team at Berkeley have been already addressed in the Ptolemy project, and for a number of others, we have ideas that we expect will lead to a solution. The Ptolemy project studies heterogeneous modeling, simulation, and design of concurrent systems. The focus is on embedded systems [Lee], particularly those that mix technologies, including for example analog and digital electronics, hardware and software, and electronics and mechanical devices. The focus is also on systems that are complex in the sense that they mix widely different operations, such as signal processing, feedback control, sequential decision making, and user interfaces.

Download

Automated ML-based sample centring for macromolecular X-ray crystallography with MXAimbot

Acta Crystallographica Section A Foundations and Advances, 2021

MXAimbot is a neural network-based tool, designed to automate the task of centering samples for m... more MXAimbot is a neural network-based tool, designed to automate the task of centering samples for macro-molecular X-ray crystallography experiments before exposing the sample to the beam. MXAimbot uses a convolutional neural network (CNN) trained on a few thousands' images from an industrial vision camera pointed at the sample to predict suitable crystal centering for subsequent data collection. The motivation for this project is that the machine vision automated sample positioning allows X-ray laboratories and synchrotron beamlines to offer a more efficient alternative for the manual centering, which is time consuming and difficult to automate with conventional image analysis, and for the X-ray mesh scan centering, which can introduce radiation damage to the crystal. MXAimbot can be used to improve results of standard LUCID loop centering for fully automated data collection in fragment-screening campaigns. No need for sample rotation should be an additional advantage. A few original approaches and CNN architectures were tested by authors in [1,2]. They were using X-ray data from mesh scans and not relying on manual annotations. Finally for a current production a simpler method inspired by a DeepCentering approach [3] from SPring-8, has been adopted. The original training dataset was manually annotated with bounding boxes around each crystal and the new CNN architecture is using the annotated data. MXAimbot can be used by other systems via a REST API. The next step for the project was including MXAimbot into MXCuBE3 -the common data acquisition framework at several European synchrotron facilities. This allows collection of anonymised datasets from the sample vision camera in the BioMAX beamline at the MAX IV synchrotron which can be further used for training and optimisation of CNNs and later be seamlessly included as an additional option in the MXCuBE3 data collection pipeline. To the authors knowledge CNNs have been implemented for crystal centering at least at two synchrotron facilities including MAX IV. So far the CNN approach has shown outstanding results in automatically positioning crystals. Work is currently underway to test and statistically compare the model predictions to the manual centerings by real users with the goal of integrating MXAimbot into the FragMAX [4] -fragment screening facility at the MAX IV sychrotron.

Download

Object-based abstract state machines

When designing and implementing complex computer sys tems powerful description techniques are nee... more When designing and implementing complex computer sys tems powerful description techniques are needed that allow the repre sentation of hardware and software components at various abstraction levels and support re nement during the development process These techniques should be formal in order to support mathematical reason ing about systems properties while at the same time they should be operational so as to facilitate comprehension and implementation of the system In this paper we propose Object based Abstract State Machines ObASM an extension of Abstract State Machines ASM ASM are an operational formalism that has been successfully applied to the description of hard ware and software systems ObASM are adding a notion of locality of state and the concept of communication between loosely coupled agents These extensions are designed to be minimal additions to the traditional ASM formulation preserving their simplicity while at the same time signi cantly enhancing their modeling powe...

Download

Aegis : Reliable Application Execution Over the Mobile Cloud

Procedia Computer Science, 2017

Abstract: With the advent of IoT and the associated variety of pervasive and context-aware applic... more Abstract: With the advent of IoT and the associated variety of pervasive and context-aware applications, there is an increasing requirement to support the execution of these applications on devices with limited processing power. This is a cost-intensive process as this usually requires the deployment of centralized computing infrastructure accessible via a cloud interface. We envision a distributed execution environment comprised of diverse computing resources as an alternative solution to this problem. This execution environment can be defined as a mobile cloud. However, the inherent unreliability of the mobile cloud requires developers to add reliability within the applications separately making the development process tedious. In this position paper, we present Aegis - a framework that provides a reliable and unified computing platform with built-in failure detection and repair mechanisms. Aegis draws upon the actor based execution model as well as stream processing applications to provide a reliable overlay over unreliable environments. Aegis also relives developers of the task of adding reliability mechanisms to applications separately.

Mobies Position Paper (DRAFT 2 Version)

Download

Design patterns in Petri net system modeling

Proceedings. Fourth IEEE International Conference on Engineering of Complex Computer Systems (Cat. No.98EX193)

Petri nets are an established and well researched means for systems modeling and simulation, but ... more Petri nets are an established and well researched means for systems modeling and simulation, but its use in the engineering community is not as widespread as the applicability of the formalism would suggest. A reason for this might lie in the fact that there is no established concept for the concise presentation of reusable Petri net design knowledge. This paper proposes Petri net design patterns as a style of presentation of such design knowledge ranging from building blocks to architectural considerations. The template for description is introduced using a number of examples taken from our design experience.

Download

Describing the Syntax and Semantics of UML Statecharts in a Heterogeneous Modelling Environment

Lecture Notes in Computer Science, 2002

In this paper UML statechart diagrams are used as an example of a generic approach to integrating... more In this paper UML statechart diagrams are used as an example of a generic approach to integrating a visual language in a heterogeneous modelling and simulation environment. A system represented in a visual language is syntactically defined as an attributed graph, with well-formedness rules specified by a set of firstorder predicates over the abstract syntax of the graph. The language semantics are specified by an Abstract State Machine (ASM) parameterized with syntacticallycorrect attributed graphs. In this paper the key issues in the definition of UML statechart semantics are highlighted.

Download

Optimization of portable parallel signal processing applications by design space exploration of dataflow programs

2011 IEEE Workshop on Signal Processing Systems (SiPS), 2011

ABSTRACT This paper describes a methodology for the optimization of portable parallel signal proc... more ABSTRACT This paper describes a methodology for the optimization of portable parallel signal processing applications specified by dataflow programs. The use of dataflow as a programming model for signal processing applications targeting parallel platforms provides an important advantage versus the traditional sequential programming paradigm: the portability of parallelism. The paper introduce a design space exploration methodology for exploring alternative implementations in which abstract traces of a program, representing the actual data dependencies of its parts, are first constructed and then analyzed to guide the refactoring and mapping and of the signal processing applications to best match its intended parallel target. The methodology is demonstrated and evaluated in an at-size case study of an MPEG-4 video decoder.

Toward efficient execution of dataflow actors

2012 Conference Record of the Forty Sixth Asilomar Conference on Signals, Systems and Computers (ASILOMAR), 2012

ABSTRACT Dataflow descriptions are a natural match to application areas such as signal processing... more ABSTRACT Dataflow descriptions are a natural match to application areas such as signal processing, cryptography, networking, image processing, and media coding. This paper addresses the problem of efficiently executing the basic elements of a dataflow program, its actors, written in a language such as MPEG&#39;s RVC-CAL. Using actor machines as an execution model for dataflow actors, we devise a metric for measuring the quality of a translation in terms of program size and execution efficiency, and then build, evaluate and compare a number of translators with each other and prior art, using MPEG reference code as a benchmark.

Profiling dataflow programs

2008 IEEE International Conference on Multimedia and Expo, 2008

Since computation in multimedia systems is often stream-oriented, it is often best represented as... more

Generalizing lookahead—behavioral prediction in distributed simulation

ACM SIGSIM Simulation Digest, 1998

is the detection of model-inherent concurrency, which is related on the prediction of the future ... more is the detection of model-inherent concurrency, which is related on the prediction of the future behavior of each logical simulation process. In this paper we present a generalization of traditional approaches to behavioral prediction which provides more accurate predictions in general and a high degree of scalability in terms of computational and communication overhead. We give a sequential implementation, prove its correctness using concepts from the analysis of real-time process networks and then show how these results can be used in distributed simulation.

Download

Partitioning and optimization of high level stream applications for multi clock domain architectures

SiPS 2013 Proceedings, 2013

ABSTRACT In this paper we propose a design methodology to partition dataflow applications on a mu... more ABSTRACT In this paper we propose a design methodology to partition dataflow applications on a multi clock domain architecture. This work shows how starting from a high level dataflow representation of a dynamic program it is possible to reduce the overall power consumption without impacting the performances. Two different approaches are illustrated, both based on the post-processing and analysis of the causation trace of a dataflow program. Methodology and experimental results are demonstrated in an at-size scenario using an MPEG-4 Simple Profile decoder.

Portable and scalable parallelism for multi-core and reconfigurable hardware using dataflow programs

Automatic queue sizing for dataflow applications

Synthesis and optimization of high-level stream programs

ABSTRACT In this paper we address the problem of translating high-level stream programs, such as ... more ABSTRACT In this paper we address the problem of translating high-level stream programs, such as those written in MPEG&#39;s RVC-CAL dataflow language, into implementations in programmable hardware. Our focus is on two aspects: sufficient language coverage to make synthesis available for a large class of programs, and methodology and tool support providing analysis and guidance to improve and optimize an initial implementation. Our main results are (1) a synthesis tool that for the first time translates a complete and unmodified MPEG reference implementation into a working hardware description, and (2) a suite of profiling and analysis tools that analyze the structure of computation weighted by data obtained from the synthesis process, and accurately pinpoint parts of the program that are targets for optimization.

Modular Analysis of Dataflow Process Networks

Lecture Notes in Computer Science, 2003

Process networks are popular for modelling distributed computing and signal processing applicatio... more Process networks are popular for modelling distributed computing and signal processing applications, and multi-processor architectures. At the architecture description level, they have the flexibility to model actual processes using various formalisms. This is especially important where the systems are composed of parts with different characteristics, e.g. control-based or dataflow-oriented. However, this heterogeneity of processes presents a challenge for the analysis of process networks. This research proposes a lightweight method for analysing properties of such networks, such as freedom from unexpected reception and deadlock. The method employs interface automata as a bridge between the architectural model and heterogeneous processes. Thus, the properties are determined by a series of small tasks at both the architecture level and the process level. This separation of concerns simplifies the handling of heterogeneous processes and alleviates the potential state space explosion problem when analysing large systems.

Download

A method for describing the syntax and semantics of UML statecharts

Software & Systems Modeling, 2004

In this article we present a method for describing the language of UML statecharts. Statecharts a... more In this article we present a method for describing the language of UML statecharts. Statecharts are syntactically defined as attributed graphs, with well-formedness rules specified by a set of first-order predicates over the abstract syntax of the graphs. The dynamic semantics of statecharts is defined by Abstract State Machines parameterized with syntactically-correct attributed graphs. The presented approach covers many important constructs of UML statecharts, including internal, completion, interlevel and compound transitions as well as history pseudostates. It also contains strategies to handle state entry/exit actions, state activities, synch states and choice pseudostates.

Download

Syntax and semantics of graphs-An approach to the specification of visual notations for discrete-event systems

Modeling a die bonder with Petri nets: a case study

IEEE Transactions on Semiconductor Manufacturing, 1998

The process of designing complex manufacturing machines involes problems of interdisciplinary com... more The process of designing complex manufacturing machines involes problems of interdisciplinary communication, design space exploration as well as design evaluation and analysis that can best be solved by integrating modeling and simulation stages into the development process. Requirements necessary for a modeling environment are: support for hierarchical and modular structuring, module reuse, con gurability of the model, excutability of the model with a notion of time, and intuitive usability. The CodeSign object-oriented, temporal Petri net formalism with its associated tool is a modeling and simulation environment to meet those requirements. A case study conducted in an industrial setting is described that demonstrates the applicability of the CodeSign approach to the modeling of semiconductor manufacturing equipment with step-wise re nement. Results are given to show the accuracy of simulation results with respect to data actually measured on the machine. The CodeSign simulation results are shown to have superior accuracy compared to the output of a previous spreadsheet model of the same machine. Manufacturing machines are systems with a high degree of complexity which is caused by the fact that they consist of strongly heterogeneous subsystems, that is, they contain distributed processors and involve several di erent design domains, such as analog and digital hardware, software, control, mechanical engineering, and, in the case of semiconductor manufacturing machines, also material science and physics. Hard real-time constraints pose additional challenges. The design process for such a system consists of development and composition of components according to the system speci cations while allowing enough exibilty to meet special customer demands with regard to con guration and usage variations, evaluating whether the design end product meets performance and reliability requirements, providing for the integration of the machine into a Computer Integrated Manufacturing environment, identifying subsystems that are critical with respect to any of the aspects above, and documentation. These tasks are complicated 3] because no person in the interdisciplinary design team knows the the system as a

Download

Jorn W Janneck

Uploads

Papers by Jorn W Janneck

Log In