TelaMalloc: Efficient On-Chip Memory Allocation for Production Machine Learning Accelerators
Martin Maas, Ulysse Beaugnon, Arun Chauhan, Berkin Ilbeyi · 2022
Memory buffer allocation for on-chip memories is a major challenge in modern machine learning systems that target ML accelerators. In interactive systems such as mobile phones, it is on the critical path of launching ML-enabled applications. In data centers, it is part of complex optimization loops that run many times and are the limiting factor for the quality of compilation results.