Section: New Results

Programming Languages for Distributed Systems

One of the objectives of the Scale team is to design programming models easing the development and safe execution of distributed systems. This section describes our results in this direction.

Multi-active Objects

Participants : Ludovic Henrio, Fabrice Huet, Justine Rochas, Vincenzo Mastandrea.

The active object programming model is particularly adapted to easily program distributed objects: it separates objects into several activities, each manipulated by a single thread, preventing data races. However, this programming model has its limitations in terms of expressiveness – risk of deadlocks – and of efficiency on multicore machines. We proposed to extend active objects with local multi-threading. We rely on declarative annotations for expressing potential concurrency between requests, allowing easy and high-level expression of concurrency. This year we realized the following:

• We published the extension of multi-active objects to support scheduling and thread limitation [12] .

• We developed a compiler from ABS language into ProActive multi-active objects. This translation can be generalised to many other active object languages. This work has been published as a research report [25] , and is under submission to a conference.

• We started to work on static detection of deadlocks for multi-active object. This is the work of Vincenzo Mastandrea who is starting a Labex PhD in collaboration with the FOCUS EPI (Univ of Bologna).

• Extensive use of multiactive objects in our CAN P2P network and implementation of usecases [2] .

• We formalised in Isabelle/HOL a first version of the semantics of multiactive objects. This work was done in collaboration with Florian Kammuller

We plan to continue to improve the model, especially about compile-time checking of annotations and about fault tolerance of multiactive objects.

Autonomic Monitoring and Management of Components

Participants : Françoise Baude, Ludovic Henrio.

We have completed the design of a framework for autonomic monitoring and management of component-based applications. We have provided an implementation using GCM/ProActive taking advantage of the possibility of adding components in the membrane. The framework for autonomic computing allows the designer to describe in a separate way each phase of the MAPE autonomic control loop (Monitoring, Analysis, Planning, and Execution), and to plug them or unplug them dynamically.

• This year, we published a journal paper summarising our approach in the GCM/ProActive framework and our contribution on componentised membranes for autonomic computing [3] .

• We also improved, in the context of the SCADA associate team and during the internship of Matias Ibañez, the support for autonomic components, providing all the architecture and API so that the programmer of autonomic aspect can do them in a DSL reconfiguration language, called GCMScript. This was implemented and experimented, a publication is under submission on this work.

Algorithmic skeletons

Participant : Ludovic Henrio.

In the context of the SCADA associated team, we worked on the algorithmic skeleton programming model. The structured parallelism approach (skeletons) takes advantage of common patterns used in parallel and distributed applications. The skeleton paradigm separates concerns: the distribution aspect can be considered separately from the functional aspect of an application. In the previous year we designed the possibility for a skeleton to output events, which increases the control and monitoring capabilities. This year we published our previous results in [14] and realised additional steps:

• Study of different ways to predict the execution time for a skeleton, inspired from simple statistic functions. This improvement together with the distributed execution of skeletons should allow us to publish a journal paper on this subject in 2015

Optimization of data transfer in event-based programming models

Participants : Iyad Alshabani, Françoise Baude, Laurent Pellegrino.

In [6] , we extended a previous work with conceptual and experimental performance evaluations. This previous and collaborative work [1] developed an innovative approach of “lazy copy and transfer” of the data parts of event objects exchanged by peers in the context of event-driven architecture applications.

While event notifications are routed in a conventional manner through an event service, data parts of the events are directly and transparently transferred from publishers to subscribers. The theoretical analysis shows that we can reduce the average event delivery time by half, compared to a conventional approach requiring the full mediation of the event service. The experimental analysis confirms that the proposed approach outperforms the conventional one (both for throughput and delivery time) even though the middleware overhead, introduced by the specific adopted model, slightly reduces the expected benefits.

Behavioural Semantics

Participants : Ludovic Henrio, Eric Madelaine, Min Zhang.

We have studied Parameterised Networks of Automata (pNets) from a theoretical perspective. We started with some 'pragmatic' expressiveness of the pNets formalism, showing how to express a wide range of classical constructs of (value-passing) process calculi, but also complex interaction patterns used in modern distributed systems. Our framework can model full systems, using (closed) hierarchies of pNets; we can also build (open) pNet systems expressing composition operators. Concerning more fundamental aspects, we defined a strong bisimulation theory specifically for the pNet model, proved its properties, and illustrated it on some examples. One of the original aspects of the approach is to relate the compositional nature of pNets with the notion of bisimulation; this was exemplified by studying the properties of a flattening operator for pNets. This work has been accepted for publication at PDP'2015 ([24] ).

A Time-sensitive Heterogeneous Behavioural Model

Participants : Eric Madelaine, Yanwen Chen.

This work concludes the PhD research of Yanwen Chen, targetting a timed-sensitive extension of the pNets model with logical clocks inherited from the CCSL language. The main results of this year are: 1) a new notion of Time Specification (TS), used to handle the abstract properties of each level of processes in a pNet structure, 2) algorithms to compute such TSs for basic parameterized and timed processes, and from composition of timed-pNets, 3) conditions for checking the compatibility of composition, 4) a use-case from the area of intelligent transportation systems, illustrating the whole chain of modeling, upto a symbolic simulation of the full composed system, with the TimeSquare tool. This work was published as [4] , [23] , and in the PhD thesis of Y. Chen, defended on 2014, Nov. 30th.

Structure and structural correctness for GCM components

Participants : Ludovic Henrio, Oleksandra Kulankhina, Eric Madelaine.

We have defined a set of rules characterizing the well-formed composition of components in order to guarantee their safe deployment and execution. This work focuses on the structural aspects of component composition; it puts together most of the concepts common to many component models, but never formalized as a whole. Our formalization characterizes correct component architectures made of functional and non-functional aspects, both structured as component assemblies. So-called 'Interceptor chains' can be used for a safe and controlled interaction between the two aspects. Our well-formed components guarantee a set of properties ensuring that the deployed component system has a correct architecture and can run safely. Those definitions constitute the formal basis for VerCors tool. This work was done in the context of O. Kulankhina phd research, and in collaboration with Dongqian Liu (ECNU Shanghai), as part of the Associated Team DAESD.