Non-Stop Microprocessor for Fault-Tolerant Real-Time Systems

Abstract

It is very important to design an embedded real-time system as a fault-tolerant system to ensure dependability. In particular, when a power failure occurs, restart processing after power restoration is required in a real-time system using a conventional processor. Even if power is restored quickly, the restart process takes a long time and causes deadline misses. In order to design a fault-tolerant real-time system, it is necessary to have a processor that can resume operation in a short time immediately after power is restored, even if a power failure occurs at any time. Since current embedded real-time systems are required to execute many tasks, high schedulability for high throughput is also important. This paper proposes a non-stop microprocessor architecture to achieve a fault-tolerant real-time system. The non-stop microprocessor is designed so as to resume normal operation even if a power failure occurs at any time, to achieve little performance degradation for high schedulability even if checkpoint creations and restorations are performed many times, to control flexibly non-volatile devices through software configuration, and to ensure data consistency no matter when a checkpoint restoration is performed. The evaluation shows that the non-stop microprocessor can restore a checkpoint within 5µsec and almost hide the overhead of checkpoint creations. The non-stop microprocessor with such capabilities will be an essential component of a fault-tolerant real-time system with high schedulability.

A Learning-based Control Scheme for MTJ-based Non-volatile Flip-Flops

“…Since a VR-NVFF is designed based on an SSR-NVFF, the parameters in Table 1 refer to those of an SSR-NVFF [9]. We assume the energy consumption of a verify operation (E V ) is the same as that of a restore operation because a verify operation is a restore operation that restores the non-volatile circuit value to the balloon circuit [20]. Note that I P→AP 1.2V , I AP→P 1.2V , and E V 1.2V in Table 1 are measured where the voltage is 1.2 V. Table 2 shows the other evaluation parameters.…”

“…Therefore, we employ NVFFs with a verify operation that verify the success or failure of a store operation. Furthermore, we assume a non-volatile checkpoint is created by multi-step store (MSS) checkpointing [20]. MSS checkpointing ensures the success of a checkpoint creation by repeating store operations until all NVFFs succeed in a store operation.…”

“…Our second objective is to reduce the energy consumption of a checkpoint creation as much as possible while meeting the first objective. Typically, MSS checkpointing shortens the duration of the first store operation so that more NVFFs succeed in the store operation with less energy and lengthens the durations of the second and subsequent store operations so that all NVFFs succeed in the store operations with a fewer number of store operations [15], [20]. MSS checkpointing repeats store operations until all NVFFs succeed in a store operation.…”

“…Furthermore, we can find the first store operation's duration and voltage that reduce the checkpoint creation's energy consumption by calculating the energy consumption formula's partial derivative with respect to the first store operation's duration and voltage. We can calculate the energy consumption of MSS checkpointing (E MSS ) such that the limit of the retry counter is set to 0 as follows [20]:…”

“…Therefore, the sign of the objective c ⃝ 2024 Information Processing Society of Japan variables for temperature should be the same as that of the objective variables for voltage. Equations ( 17), (18), and (20) show that voltage has a more significant impact on the success rate of store operations than temperature. Therefore, the absolute value of the objective variable for temperature should be smaller than that of the objective variable for voltage.…”