5 papers
MANTA: Machine Learning Augmented Tiering Advisor
Johannes Freischuetz, Kiet Pham, Sujay Yadalam +3
Memory tiering has been used to expand memory capacity, particularly in datacenters, by combining fast DRAM with slower tiers, including CXL-attached memory. Its effectiveness depe…
Vulcan: Instance-specialized, Verifiable Systems Heuristics Through LLM-driven Search
Rohit Dwivedula, Divyanshu Saxena, Sujay Yadalam +3
Systems resource management tasks rely primarily on hand-designed heuristics. However, growing hardware heterogeneity and workload diversity require heuristics specialized to parti…
ARMS: Adaptive and Robust Memory Tiering System
Sujay Yadalam, Konstantinos Kanellis, Michael Swift +1
Memory tiering systems seek cost-effective memory scaling by adding multiple tiers of memory. For maximum performance, frequently accessed (hot) data must be placed close to the ho…
From Good to Great: Improving Memory Tiering Performance Through Parameter Tuning
Konstantinos Kanellis, Sujay Yadalam, Fanchao Chen +2
Memory tiering systems achieve memory scaling by adding multiple tiers of memory wherein different tiers have different access latencies and bandwidth. For maximum performance, fre…
Herding LLaMaS: Using LLMs as an OS Module
Aditya K Kamath, Sujay Yadalam
Computer systems are becoming increasingly heterogeneous with the emergence of new memory technologies and compute devices. GPUs alongside CPUs have become commonplace and CXL is p…