Home         Authors   Papers   Year of conference   Themes   Organizations        To MES conference

Design principles for fault-tolerant random access memory for space applications
Authors
	Volobuev S.V.
	Evdokimov A.P.
	Ryabtsev V.G.
Date of publication
	2020
DOI
	10.31114/2078-7707-2020-2-103-109
Abstract
	The main components of microprocessor control devices are storage devices that store programs and are used when executing control algorithms. However, in memory chips under the influence of static electricity, electron migration due to the high conductivity of oxides, the tunneling effect, etc. over time, failures and malfunctions can occur, which can lead to a catastrophe of critical application systems, which include protection systems for reactors of nuclear power plants, control systems for aerospace and other objects. The memory self-repair technique called BISR (Built-In Self-Repair) is implemented by the fuse burn controller directly on the chip at the manufacturer’s platforms, but automatic repair of storage devices by this technique in the event of repeated failures on the user's platforms is not possible. If the first memory module fails, the means of operating diagnostics are detected and the second memory module is connected to the control system, which is currently in working condition and stores data and / or the control program. The third module switches to data mirroring mode, and the first module switches to the self-test and automatic recovery mode. This interaction of tools such as operating, built-in test diagnostics, and automatic recovery from multiple failures increases the reliability of control microprocessor systems, which is especially important for critical applications, which include protection systems for reactors of nuclear power plants, other power facilities, space technology, marine and ground equipment vehicles, airplanes and other aircraft. Repair of control systems for unmanned aerial vehicles will be possible automatically without the participation of personnel. The multiple increases in the reliability of control computer systems, which is especially important for mission-critical systems applications, including space equipment, aircraft and other airborne devices. Repair of storage management systems for an unmanned aerial vehicle will be possible automatically without the participation of personnel. Implementation of the built-in self-test significantly reduces the cost of funds and increases the percentage of fault coverage, since test diagnostics are performed at operating frequencies and external test equipment is not required, the cost of which, as a rule, is many times higher than the cost of the memory modules themselves.
Keywords
	memory mirroring, multiple failures, redundancy, self-testing and self-repair tools.
Library reference
	Volobuev S.V., Evdokimov A.P., Ryabtsev V.G. Design principles for fault-tolerant random access memory for space applications // Problems of Perspective Micro- and Nanoelectronic Systems Development - 2020. Issue 2. P. 103-109. doi:10.31114/2078-7707-2020-2-103-109
URL of paper
	http://www.mes-conference.ru/data/year2020/pdf/D003.pdf