Experiences with Locking in a NUMA Multiprocessor Operating System Kernel

We describe the lockingarchitecture of a new operating system, HURRICANE, designed for large scale shared-memory multiprocessors. Many papers already describe kernel lockingtechniques,andsomeofthetechniquesweusehave been previouslydescribed by others. However, our work is novel in the particular combination of techniques used, as well as several of the individual techniques themselves. Moreover, it is the way the techniques work together that is the source of our performance advantages and scalability. Briefly, we use: a hybrid coarse-grain/fine-grain locking strategy that has the low latency and space overhead of a coarsegrain locking strategy while having the high concurrency of a fine-grain locking strategy; replication of data structures to increase access bandwidth and improve concurrency; aclusteredkernel that boundsthenumberofprocessors that can compete for a lock so as to reduce second order effects such as memory and interconnect contention; Distributed Locks to further reduc...

• Type: Conference
• Year: 1994
• Where: OSDI
• Authors: Ronald C. Unrau, Orran Krieger, Benjamin Gamsa, Michael Stumm
• Keywords: Coarse-grain/fine-grain Locking Strategy, Coarsegrain Locking Strategy, Operating System, OSDI 1994, Scale Shared-memory Multiprocessors