Cache-Line Transactions

Köppen, Marcel; Traue, Jana; Borchert, Christoph; Nolte, Jörg; Spinczyk, Olaf

doi:10.1145/3365137.3365396

Cited by 3 publications

(1 citation statement)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The technique selects one of copy on write and shifting in place strategies to perform data insertion or deletion at runtime thereby to enable failure resilience. In Reference [23], the authors present a transaction mechanism for small data structures, limited to the half of a cache line, to persist data in NVM. Although this mechanism cannot be used for general arbitrary applications, it is designed to be a building block of techniques that persists more complicated data structures.…”

Section: Related Workmentioning

confidence: 99%

Compiler-support for Critical Data Persistence in NVM

Elkhouly

Alshboul

Hayashi

et al. 2019

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

Non-volatile Main Memories (NVMs) offer a promising way to preserve data persistence and enable computation recovery in case of failure. While the use of NVMs can significantly reduce the overhead of failure recovery, which is the case with High-Performance Computing (HPC) kernels, rewriting existing programs or writing new applications for NVMs is non-trivial. In this article, we present a compiler-support that automatically inserts complex instructions into kernels to achieve NVM data-persistence based on a simple programmer directive. Unlike checkpointing techniques that store the whole system state, our technique only persists user-designated objects as well as some parameters required for safe recovery such as loop induction variables. Also, our technique can reduce the number of data transfer operations, because our compiler coalesces consecutive memory-persisting operations into a single memory transaction per cache line when possible. Our compiler-support is implemented in the LLVM tool-chain and introduces the necessary modifications to loop-intensive computational kernels (e.g., TMM, LU, Gauss, and FFT) to force data persistence. The experiments show that our proposed compiler-support outperforms the most recent checkpointing techniques while its performance overheads are insignificant. CCS Concepts: • Software and its engineering → Source code generation; Runtime environments; Main memory; • Hardware → Emerging languages and compilers;

show abstract

Section: Related Workmentioning

confidence: 99%