ED/sup 4/I: error detection by diverse data and duplicated instructions

Abstract. It is expected that commodity hardware is becoming less reliable because of the continuously decreasing feature sizes of integrated circuits. Nevertheless, more and more commodity hardware with insufficient error detection is used in critical applications. One possible solution is to detect hardware errors in software using arithmetic AN-codes. These codes detect hardware errors independent of the actual failure modes of the underlying hardware. However, measurements have shown that AN-codes still exhibit large rates of undetected silent data corruptions (SDC). These high rates of undetected SDCs are caused by the insufficient protection of control and data flow through AN-codes. In contrast, ANB-and ANBD-codes promise much higher error detection rates because they also detect errors in control and data flow. We present our encoding compiler that automatically applies either an AN-, ANBor ANBD-code to an application. Our error injections show that AN-, ANB-, and ANBD-codes successfully detect errors and more important that indeed ANB-and ANBD-codes reduce the SDC rate more effectively than AN-codes. The difference between ANBD-and ANB-codes is also visible but less pronounced.

“…For all approaches the error injection experiments show a non-negligible amount of undetected failures. This is even higher for [10] and [7] because the used encoding is incomplete.…”

Section: Related Workmentioning

confidence: 97%

“…Therefore, the program and the processed data are modified. ANencoding was already used by [10,7,14]. For all approaches the error injection experiments show a non-negligible amount of undetected failures.…”

Section: Related Workmentioning

confidence: 99%

ANB- and ANBDmem-Encoding: Detecting Hardware Errors in Software

Schiffel

Schmitt

Süßkraut

et al. 2010

“…[13] duplicates instruction and compares their outcomes. A similar approach is taken in ED4I [14], but the duplicated instructions do not process the original data but a k-multiple of it. Thus all results of duplicate instructions have to be k-multiples of the original results.…”

Section: Related Workmentioning

confidence: 99%

Software Encoded Processing: Building Dependable Systems with Commodity Hardware

Wappler

Fetzer

2007

Abstract. In future, the decreasing feature size and the reduced power supply will make it much more difficult to built reliable microprocessors. Economic pressure will most likely result in the reliability of microprocessors being tuned for the commodity market. In the dependability domain we expect the continued spreading of mixed-mode computing systems, i.e., systems that execute both critical and non-critical functionality. To permit the efficient execution of non-critical applications and the correct execution of critical applications, we introduce the concept of Software Encoded Processing (SEP). SEP enforces a crash failure semantics of the underlying CPU. It does not require the source code of encoded programs and provides probabilistic guarantees. To achieve this, arithmetic codes and signatures are used to detect corrupted data and faulty executions of programs.

“…In that case, the program and the processed data are modified. ED4I [20], for example, duplicates instructions but the duplicated instructions do not process the original data but a k-multiple of it, which is a so-called AN-code. All results of duplicate instructions have to be k-multiples of the original results.…”

Section: Related Workmentioning

confidence: 99%

“…3). In contrast to similar previous approaches such as [20,10,34], we encode the whole application with the same powerful code. This includes memory, logical operations, and handling of external functions whose source code is not available for encoding.…”

Section: Introductionmentioning

confidence: 99%

AN-Encoding Compiler: Building Safety-Critical Systems with Commodity Hardware

Fetzer

Schiffel

Süßkraut

2009

Abstract. In the future, we expect commodity hardware to be used in safety-critical applications. However, in the future commodity hardware is expected to become less reliable and more susceptible to soft errors because of decreasing feature size and reduced power supply. Thus, software-implemented approaches to deal with unreliable hardware will be needed. To simplify the handling of value failures, we provide failure virtualization in the sense that we transform arbitrary value failures caused by erroneous execution into fail-stop failures. The latter ones are easier to handle. Therefore, we use the arithmetic AN-code because it provides very good error detection capabilities. Arithmetic codes are suitable for the protection of commodity hardware because guarantees can be provided independent of the executing hardware. This paper presents the encoding compiler EC-AN which applies AN-encoding to arbitrary programs. According to our knowledge, this is the first in software implemented complete AN-encoding. Former encoding compilers either encode only small parts of applications or trade-off safety to enable complete AN-encoding.