Authors: Benedek Horváth(IncQuery Labs cPlc., Johannes Kepler University Linz, Linz, Austria), Ákos Horváth (IncQuery Labs cPlc.), Manuel Wimmer (Johannes Kepler University Linz, Linz, Austria)
Read the research here: https://dl.acm.org/doi/10.1145/3417990.3420199
Booking open Available Pune Call Girls Pargaon 6297143586 Call Hot Indian Gi...
Towards the Next Generation of Reactive Model Transformations on Low-Code Platforms: Three Research Lines
1. Towards the Next Generation of Reactive Model Transformations
on Low-Code Platforms:
Three Research Lines
Benedek Horváth1,2, Ákos Horváth1, Manuel Wimmer2
1 IncQuery Labs cPlc., Budapest, Hungary
2 Johannes Kepler University Linz, Linz, Austria
Contact: Benedek.Horvath@incquerylabs.com
2. Motivation
• Next generation of Model-Based Systems Engineering (MBSE) tools
• Adopt benefits of LCDPs for MBSE
• Challenges in the advancement 2
Low-Code
Engineering Platform
Challenges
Desktop-oriented
MBSE tools
Benefits of LCDPs
Collaborative
platform
Visual
diagrams
Domain-
specific editors
Cloud
deployment
Model
size
Number
of users
Scalability
Productivity
Model
transformation
and platform
characteristics
3. Transformation
engine
Query engine
Persistent
index
In-memory
index
Model ManagementService
LCEP architecture
Reporting
Model analysis,
model checking
Collaborative
editor
Low-Code Engineering Platform
Model
repository
External
tools
3
C1: Number of
users
Shared
resources,
isolation
C2: Model size
Sheer size,
multiple
revisions
C3: Model
transformation
and platform
characteristics
KPIs
4. Research lines
4
RL2: Parallel Reactive
Model Transformations
RL1: Multi-tenant Model
Transformations
RL3: Multi-tenant,
Reactive Model
Transformation
Benchmark
needs
evaluates
evaluates
Research lines
Challenges
addressesaddresses addresses
C2: Model size
C1: Number of
users
C3: Model
transformation
and platform
characteristics
5. RL1: Multi-tenant Model Transformations
5
DESKTOP WORLD
CLOUD WORLD
Model Transformation
Engine
<<run by>>
<<run by>>
TE2
TE1
LCEP
<<include>>
model
transformation
model
manipulation
Source
Model
Target
Model
Model Management Service
<<manage>> <<manage>>
Model Management
Engine
Model
TE1
Model
TE2
6. RL1: Multi-tenant Model Transformations
• Goal: Integration of LCEP with Model Management Service
• tenant-isolated or dedicated component patterns [20]
• Challenges:
• Tenant isolation
• Optimized memory access and model processing (cloud resources can be expensive)
• Short response time 6
<<run by>>
<<run by>>
TE2
TE1
LCEP
<<include>>
model
transformation
model
manipulation
Model Management Service
<<manage>> <<manage>>
Model Management
Engine
Model
TE1
Model
TE2
7. RL2: Parallel Reactive Model Transformations
• Combination of transformation approaches is barely discovered
• Incremental + lazy [36], incremental + reactive [5], incremental + parallel
[7]
• Parallel extension of Event-driven Virtual Machine (EVM) in Viatra [5]
7
Model change
Event-driven
Virtual Machine
Rule
specifications
Event
Query result
update
8. RL2: Parallel Reactive Model Transformations
• Goal: task-parallel execution mode for EVM
• Distribute rule specifications for EVM instances
• Challenges:
• Rule distribution, dependencies, scheduling
• High-throughput concurrent model access
• Model consistency
• Application area:
• Derive multiple formal models from the engineering models
8
TM2 1
model change
query
match set
query
match set
TM1 TM2 2
Source
Model
Target
Model1
Incremental
Query Engine
Target
Model2
EVM1 EVM2 EVM3
9. RL3: Multi-tenant, Reactive Model Transformation Benchmark
• Goal: benchmark reactive model transformations on multi-tenant platforms
• Challenges:
• KPI identification to compare the engines (multi-tenancy, reactivity)
• Reactive scenarios: atomic operations, complex stories, execution 9
Transformation
description
Source
model n
Source
model n
Source
model
Benchmarkresult
Benchmark
orchestrator
Transformation
engine
Reactive
scenario
10. Conclusions and future work
10
• Identified scalability and productivity challenges in LCDPs and LCEPs
• Mapped challenges to research lines:
• RL1: Multi-tenant Model Transformations
• RL2: Parallel Reactive Model Transformations
• RL3: Multi-tenant, Reactive Model Transformation Benchmark
• Implementations to be integrated to IncQuery Server [23], to enhance it into a
multi-tenant, collaborative engineering platform over cloud-based model
repositories.
This work is funded by the European Union’s Horizon 2020 research and innovation programme
under the Marie Skłodowska-Curie grant agreement No 813884.
The authors are grateful for the valuable feedback of Dániel Varró, István Ráth and the anonymous
reviewers about the paper.
11. References
11
[5] Gábor Bergmann, István Dávid, Ábel Hegedüs, Ákos Horváth, István Ráth, Zoltán Ujhelyi,
and Dániel Varró. 2015. Viatra 3: A Reactive Model Transformation Platform. In Proc of the
8th International Conference on the Theory and Practice of Model Transformations (LNCS,
Vol. 9152). Springer, 101–110.
[7] Gábor Bergmann, István Ráth, and Dániel Varró. 2009. Parallelization of Graph
Transformation Based on Incremental Pattern Matching. ECEASST 18 (2009).
[12] Hong Cai, Ning Wang, and Ming Jun Zhou. 2010. A Transparent Approach of Enabling
SaaS Multi-tenancy in the Cloud. In Proc of the 6th World Congress on Services. IEEE
Computer Society, 40–47.
[20] Christoph Fehling, Frank Leymann, Ralph Retter, Walter Schupeck, and Peter Arbitter.
2014. Cloud Computing Patterns - Fundamentals to Design, Build, and Manage Cloud
Applications. Springer.
[23] Ábel Hegedüs, Gábor Bergmann, Csaba Debreceni, Ákos Horváth, Péter Lunk, Ákos
Menyhért, István Papp, Dániel Varró, Tomas Vileiniskis, and István Ráth. 2018. Incquery
server for teamwork cloud: scalable query evaluationover collaborativemodel repositories.
In MODELS. ACM, 27–31.
12. References
12
[24] Kai Hu, Lei Lei, and Wei-Tek Tsai. 2016. Multi-tenant Verification-as-a-Service (VaaS) in a
cloud. Simulation Modelling Practice and Theory 60 (2016), 122–143.
[33] Ralph Mietzner, Tobias Unger, Robert Titze, and Frank Leymann. 2009. Combining
Different Multi-tenancy Patterns in Service-Oriented Applications. In Proc. of the 13th
International Enterprise Distributed Object Computing Conference. IEEE, 131–140.
[36] Salvador Martínez Perez, Massimo Tisi, and Rémi Douence. 2017. Reactive model
transformation with ATL. Science of Computer Programming 136 (2017), 1–16.
[43] Gábor Szárnyas, Benedek Izsó, István Ráth, and Dániel Varró. 2018. The Train
Benchmark: cross-technology performance evaluation of continuous model queries. SoSyM
17, 4 (2018), 1365–1393.
13. Motivating example from MBSE
• Automated methods to ensure model correctness
• Correctness: i.e. syntax, structure, behavior, deployment
• Check behavioral correctness: simulation, formal verification
13
Horváth et al. Model Checking as a Service: Towards Pragmatic Hidden Formal
Methods. In OpenMBEE ’20: Workshop on Open Model Based Engineering
Environment. https://doi.org/10.1145/1122445.1122456
LCEP
Static checks
Properties
Formal model Model
checker
14. Related work: Model transformation and query approaches
• Combination of approaches:
• Incremental and lazy: Perez et al. Reactive model transformation with ATL [36]
• Incremental and reactive: Bergmann et al. Viatra 3: A Reactive Model Transformation
Platform [5]
• Incremental and parallel: Bergmann et al. Parallelization of Graph Transformation Based
on Incremental Pattern Matching [7]
• Combination of parallel and reactive is not exploited yet 14
15. Related work: multi-tenant architectures
• Widely researched in SaaS applications [12, 20, 33]
• Hu et al. Multi-tenant Verification-as-a-Service (VaaS) in a cloud [24]
• Research opportunity: specialization for MDE and model transformations
(MT)
15
16. Related work: MT performance evaluation
• Custom cases → difficult to compare
• TTC, AGTIVE, GraBats:
• Do not focus on reactive transformations nor on multi-tenancy
• Szárnyas et al. The Train Benchmark: cross-technology performance
evaluation of continuous model queries [43]
• Room for benchmark on parallel reactive MTs on multi-tenant platforms
16