MH-QEMU: Memory-State-Aware Fault Injection Platform

Abstract

As we move towards higher-density, larger-scale, and lowerpower computing hardware, new types of failures are being experienced with increasing frequency. Hardware designed for the post-Moore generation are also bringing about novel resiliency challenges. In order to improve the efficiency of resiliency methods, fault injection plays an important role in understanding how errors affect the OS and application. Memory-state-aware fault injection, in particular, can be used to investigate the memory-related faults caused by using current and future hardware under extreme conditions and assess the costs/benefit trade-off of resiliency methods. We introduce MH-QEMU, a memorystate-aware fault injection platform implemented by extending a virtual machine (VM) to intercepting memory accesses. MH-QEMU supports collecting the physical and virtual addresses of memory accesses and defining appropriate injections condition using the collected information. MH-QEMU incurs a 3.4× overhead, and we demonstrate how row-hammer faults can be injected using MH-QEMU to analyzing the resiliency modified NPB CG's algorithm.

A Survey of QEMU-Based Fault Injection Tools & Techniques for Emulating Physical Faults

“…MH-QEMU, a memory fault injector that is memory state-aware and implemented by extending a VM to intercept memory accesses, is introduced in [47]. MH-QEMU follows the same method as F-SEFI, [44], the only difference between MH-QEMU and F-SEFI is that MH-QEMU focuses on memory module faults that define memory state and memory access patterns.…”

“…With all the functionalities of FI in place, MH-QEMU performance evaluation shows that it is up to 3.4 times slower than a typical native QEMU implementation. Narrowing down the monitored memory region can reduce the overhead obtained [47]. analyses them to create an appropriate fault injection plan, and applies it to target VM memory which is done from the host to avoid side effects to the target system'' [47].…”

“…Narrowing down the monitored memory region can reduce the overhead obtained [47]. analyses them to create an appropriate fault injection plan, and applies it to target VM memory which is done from the host to avoid side effects to the target system'' [47]. [46] is another fault injector that uses QEMU for checking the reliability of the memory, hard disk, and network controller system components.…”

“…• Memory -The memory cells such as RAM and related parts are tested [44], [45], [46], [47], [50], [51], [54], [55], [63]. • Registers -General purpose and special purpose registers are tested in [44], [45], [48], [49], [50], [51], [53], [54], [58], and [66].…”

“…• Transient [39], [48], [49], [53], [58], [63] • Intermittent [48], [58], [63] • Permanent [45], [48], [49], [50], [54], [58], [63], [66] faults. There are faults such as packet drop and packet modification [55], memory-state-aware faults [47], and faulty instructions [44] used by different tools.…”