Si-Kintsugi: Towards Recovering Golden-Like Performance of Defective Many-Core Spatial Architectures for AI
Open Access
- 28 October 2023
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 972-985
Abstract
The growing demand for higher compute and memory capacity driven by artificial intelligence (AI) applications pushes higher core counts in modern systems. Many-core architectures exhibiting spatial interconnects with high on-chip bandwidth are ideal for these workloads due to their data movement flexibility and sheer parallelism. However, the size of such platforms makes them particularly susceptible to manufacturing defects, prompting a need for designs and mechanisms that improve yield. Despite these techniques, nonfunctional cores and links are unavoidable. Although prior works address defective cores by disabling them and only scheduling workload to functional ones, communication latency through spatial interconnects is tightly associated with the locations of defective cores and cores with assigned work. Based on this observation, we present Si-Kintsugi, a defect-aware workload scheduling framework for spatial architectures with mesh topology. First, we design a novel and generalizable workload mapping representation and cost function that integrates defect pattern information. The mapping representation is formed into a 1D vector with simple constraints, making it an ideal candidate for open source heuristic-based optimization algorithms. After a communication latency optimized workload mapping is found, dataflow between the mapped cores is automatically generated to balance communication and computation cost. Si-Kintsugi is extensively evaluated on various workloads (i.e., BERT, ResNet, GEMM) across a wide range of defect patterns and rates. Experiment results show that Si-Kintsugi generates a workload schedule that is on average 1.34 × faster than the industry standard layer-pipelined schedule on defective platforms.Keywords
Funding Information
- National Science Foundation (CNS-2112562)
- Army Research Office (W911NF-19-2-0107)
This publication has 42 references indexed in Scilit:
- PPB: Partially-working processors binning for maximizing wafer utilization2015
- Wafer Map Failure Pattern Recognition and Similarity Ranking for Large-Scale Data SetsIEEE Transactions on Semiconductor Manufacturing, 2014
- Versatile, scalable, and accurate simulation of distributed applications and platformsJournal of Parallel and Distributed Computing, 2014
- Maximizing Yield per Area of Highly Parallel CMPs Using Hardware RedundancyIEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2014
- A survey on application mapping strategies for Network-on-Chip designJournal of Systems Architecture, 2013
- Parallel Tree Reduction on MapReduceProcedia Computer Science, 2012
- Defect and error tolerance in the presence of massive numbers of defectsIEEE Design & Test of Computers, 2004
- ADRES: An Architecture with Tightly Coupled VLIW Processor and Coarse-Grained Reconfigurable Matrix2003
- Defect tolerance in VLSI circuits: techniques and yield analysisProceedings of the IEEE, 1998
- Simulated annealing1995