Compiler-Directed Soft Error Resilience for Lightweight GPU Register File Protection
This paper presents Penny, a compiler-directed resilience scheme for protecting GPU register files (RF) against soft errors. Penny replaces the conventional error correction code (ECC) based RF protection by using less expensive error detection code (EDC) along with idempotence based recovery. Compared to the ECC protection, Penny can achieve either the same level of RF resilience yet with significantly lower hardware costs or stronger resilience using the same ECC due to its ability to detect multi-bit errors when it is used solely for detection. In particular, to address the lack of store buffers in GPUs, which causes both checkpoint storage overwriting and the high cost of checkpointing stores, Penny provides several compiler optimizations such as storage coloring and checkpoint pruning. Across 25 benchmarks, Penny causes only $\approx$3% run-time overhead on average.
Wed 17 JunDisplayed time zone: Pacific Time (US & Canada) change
06:20 - 07:40 | Networking and HardwarePLDI Research Papers at PLDI Research Papers live stream Chair(s): Luís Pina University of Illinois at Chicago | ||
06:20 20mTalk | NV: An Intermediate Language for Verification of Network Control Planes PLDI Research Papers Nick Giannarakis Princeton University, USA, Devon Loehr Princeton University, USA, Ryan Beckett Microsoft Research, USA, David Walker Princeton University, USA | ||
06:40 20mTalk | Compiler-Directed Soft Error Resilience for Lightweight GPU Register File Protection PLDI Research Papers Hongjune Kim Seoul National University, South Korea, Jianping Zeng Purdue University, USA, Qingrui Liu Virginia Tech, USA, Mohammad Abdel-Majeed University of Jordan, Jordan, Jaejin Lee Seoul National University, South Korea, Changhee Jung Purdue University, USA | ||
07:00 20mTalk | Adaptive Low-Overhead Scheduling for Periodic and Reactive Intermittent Execution PLDI Research Papers | ||
07:20 20mTalk | Detecting Network Load Violations for Distributed Control Planes PLDI Research Papers Kausik Subramanian University of Wisconsin-Madison, USA, Anubhavnidhi Abhashkumar University of Wisconsin-Madison, USA, Loris D'Antoni University of Wisconsin-Madison, USA, Aditya Akella University of Wisconsin-Madison, USA |