Section: New Results

Language Based Fault-Tolerance

Participants : Pascal Fradet, Alain Girault, Yoann Geoffroy, Gregor Goessler, Jean-Bernard Stefani, Martin Vassor, Athena Abdi.

Fault Ascription in Concurrent Systems

The failure of one component may entail a cascade of failures in other components; several components may also fail independently. In such cases, elucidating the exact scenario that led to the failure is a complex and tedious task that requires significant expertise.

The notion of causality (did an event $e$ cause an event $e^{\text{'}}$?) has been studied in many disciplines, including philosophy, logic, statistics, and law. The definitions of causality studied in these disciplines usually amount to variants of the counterfactual test “$e$ is a cause of $e^{\text{'}}$ if both $e$ and $e^{\text{'}}$ have occurred, and in a world that is as close as possible to the actual world but where $e$ does not occur, $e^{\text{'}}$ does not occur either”. In computer science, almost all definitions of logical causality — including the landmark definition of [63] and its derivatives — rely on a causal model that may not be known, for instance in presence of black-box components. For such systems, we have been developing a framework for blaming that helps us establish the causal relationship between component failures and system failures, given an observed system execution trace. The analysis is based on a formalization of counterfactual reasoning [7].

In his PhD thesis, Yoann Geoffroy proposed a generalization of our fault ascription technique to systems composed of black-box and white-box components. For the latter a faithful behavioral model is given but no specification. The approach leverages results from game theory and discrete controller synthesis to define several notions of causality.

We are currently working on an instantiation of our general semantic framework for fault ascription in  [60] to acyclic models of computation, in order to compare our approach with the standard definition of actual causality proposed by Halpern and Pearl.

Tradeoff exploration between energy consumption and execution time

We have continued our work on multi-criteria scheduling, in two directions. First, in the context of dynamic applications that are launched and terminated on an embedded homogeneous multi-core chip, under execution time and energy consumption constraints, we have proposed a two layer adaptive scheduling method. In the first layer, each application (represented as a DAG of tasks) is scheduled statically on subsets of cores: 2 cores, 3 cores, 4 cores, and so on. For each size of these sets (2, 3, 4, ...), there may be only one topology or several topologies. For instance, for 2 or 3 cores there is only one topology (a “line”), while for 4 cores there are three distinct topologies (“line”, “square”, and “T shape”). Moreover, for each topology, we generate statically several schedules, each one subject to a different total energy consumption constraint, and consequently with a different Worst-Case Reaction Time (WCRT). Coping with the energy consumption constraints is achieved thanks to Dynamic Frequency and Voltage Scaling (DVFS). In the second layer, we use these pre-generated static schedules to reconfigure dynamically the applications running on the multi-core each time a new application is launched or an existing one is stopped. The goal of the second layer is to perform a dynamic global optimization of the configuration, such that each running application meets a pre-defined quality-of-service constraint (translated into an upper bound on its WCRT) and such that the total energy consumption be minimized. For this, we (i) allocate a sufficient number of cores to each active application, (ii) allocate the unassigned cores to the applications yielding the largest gain in energy, and (iii) choose for each application the best topology for its subset of cores (i.e., better than the by default “line” topology). This is a joint work with Ismail Assayad (U. Casablanca, Morocco) who visited the team in September 2015.

Second, in the context of a static application (again represented a DAG of tasks) running on an homogeneous multi-core chip, we have worked on the static scheduling minimizing the WCRT of the application under the multiple constraints that the reliability, the power consumption, and the temperature remain below some given thresholds. There are multiple difficulties: (i) the reliability is not an invariant measure w.r.t. time, which makes it impossible to use backtrack-free scheduling algorithms such as list scheduling [33]; to overcome this, we adopt instead the Global System Failure Rate (GSFR) as a measure of the system's reliability, which is invariant with time [57]; (ii) keeping the power consumption under a given threshold requires to lower the voltage and frequency, but this has a negative impact both on the WCRT and on the GSFR; keeping the GSFR below a given threshold requires to replicate the tasks on multiple cores, but this has a negative impact both on the WCRT, on the power consumption, and on the temperature; (iii) keeping the temperature below a given threshold is even more difficult because the temperature continues to increase even after the activity stops, so each scheduling decision must be assessed not based on the current state of the chip (i.e., the temperature of each core) but on the state of the chip at the end of the candidate task, and cooling slacks must be inserted. We have proposed a multi-criteria scheduling heuristics to address these challenges. It produces a static schedule of the given application graph and the given architecture description, such that the GSFR, power, and temperature thresholds are satisfied, and such that the execution time is minimized. We then combine our heuristic with a variant of the $\epsilon$-constraint method [62] in order to produce, for a given application graph and a given architecture description, its entire Pareto front in the 4D space (exec. time, GSFR, power, temp.). This is a joint work with Athena Abdi and Hamid Zarandi from Amirkabir U., Iran, who have visited the team in 2016.

Automatic transformations for fault tolerant circuits

In the past years, we have studied the implementation of specific fault tolerance techniques in real-time embedded systems using program transformation [1]. We are now investigating the use of automatic transformations to ensure fault-tolerance properties in digital circuits. To this aim, we consider program transformations for hardware description languages (HDL). We consider both single-event upsets (SEU) and single-event transients (SET) and fault models of the form “at most 1 SEU or SET within $n$ clock cycles”.

We have expressed several variants of triple modular redundancy (TMR) as program transformations. We have proposed a verification-based approach to minimize the number of voters in TMR [25]. Our technique guarantees that the resulting circuit (i) is fault tolerant to the soft-errors defined by the fault model and (ii) is functionally equivalent to the initial one. Our approach operates at the logic level and takes into account the input and output interface specifications of the circuit. Its implementation makes use of graph traversal algorithms, fixed-point iterations, and BDDs. Experimental results on the ITC’99 benchmark suite indicate that our method significantly decreases the number of inserted voters which entails a hardware reduction of up to $55\%$ and a clock frequency increase of up to $35\%$ compared to full TMR. We address scalability issues arising from formal verification with approximations and assess their efficiency and precision. As our experiments show, if the SEU fault-model is replaced with the stricter fault-model of SET, it has a minor impact on the number of removed voters. On the other hand, BDD-based modeling of SET effects represents a more complex task than the modeling of an SEU as a bit-flip. We propose solutions for this task and explain the nature of encountered problems. We discuss scalability issues arising from formal verification with approximations and assess their efficiency and precision.

Concurrent flexible reversibility

Reversible concurrent models of computation provide natively what appears to be very fine-grained checkpoint and recovery capabilities. We have made this intuition clear by formally comparing a distributed algorithm for checkpointing and recovery based on causal information, and the distributed backtracking algorithm that lies at the heart of our reversible higher-order pi-calculus. We have shown that (a variant of) the reversible higher-order calculus with explicit rollback can faithfully encode a distributed causal checkpoint and recovery algorithm. The reverse is also true but under precise conditions, which restrict the ability to rollback a computation to an identified checkpoint. This work has currently not been published.