An error-detection and self-repairing method for dynamically and partially reconfigurable systems

Abstract

Reconfigurable systems are gaining an increasing interest in the domain of safety-critical applications, for example in space and avionic applications. In fact, the capability of reconfiguring the system during run-time execution and the high computational power of modern Field Programmable Gate Arrays (FPGAs) makes these devices suitable for data processing. Moreover, such systems must also guarantee the abilities of selfawareness, self-diagnosis and self-repair in order to cope with errors due to the harsh conditions typically existing in some environments. In this paper we propose a self-repairing method for partially and dynamically reconfigurable systems applied at a fine-grain granularity level. Our method is able to recover and correct errors using the run-time partial reconfiguration capabilities offered by modern SRAM-based FPGAs. Fault injection experiments have been executed on a dynamically reconfigurable system embedding a number of benchmark circuits. Results demonstrate that the method can achieve full detection of single and multiple errors, while significantly improving the system availability with respect to traditional error detection and correction methods.

Low Cost Dynamic Scrubbing for Real-Time Systems

“…The basic idea behind shifted scrubbing is that, by starting the scrubbing operation at an appropriately chosen frame, the repair time will be smaller than that of the standard technique. The approach herein proposed attempts to dynamically choose an improved starting frame without the need for costly fine-grained checkers as in [6,7,11]. To illustrate the concept of shift scrubbing, the following histogram for an arbitrary signature of the pdc benchmark circuit is shown (Fig.…”

“…The authors of [7] make use of DMR and divide the FPGA in a static and a dynamic region. The static region is assumed to be protected by TMR.…”

“…With this information, we anticipate the best scrubbing starting position as to maximize the repair probability given a fixed time, depending on the slack time of the system, as not to miss those deadlines even in the presence of an error. The use of CG-DMR is a significant change from previous approaches such as [6,7] which artificially introduce numerous comparators to obtain additional diagnosis information, increasing the critical path delay and thus reducing the circuit's performance and the repair time available, as will be explained in Sect. 3.1.…”

Exploiting Partial Reconfiguration through PCIe for a Microphone Array Network Emulator

“…An example is presented in [8], where different image processing operations are switched in runtime to more power-efficient modes. The runtime management of the FPGA's resources through PR allows self-adaptive or selfrepairing systems such as the one presented in [9]. Further examples of how PR improves the resource utilization and increases the flexibility of the system are detailed in [10].…”

“…for ∈ RP do (8) for ∈ Size( ) do (9) if Config( temp ( )) = Config( ( )) then (10) node [ ] ← InsertConfigInRP( ( ), ); (11) node [ ] ← MarkAsConfigured(); (12) ← RemoveConfigfromList( , ); (13) break; (14) end (15) end (16) end (17) if…”