A case for NUMA-aware contention management on multicore systems
By Sergey Blagodurov, Sergey Zhuravlev, Mohammad Dashti, Alexandra Fedorova. International Conference on Parallel Architectures and Compilation Techniques, 2010
306 citations recorded by the source corpus.
On multicore systems contention for shared resources occurs when memory-intensive threads are co-scheduled on cores that share parts of the memory hierarchy, such as lastlevel caches and memory controllers. Previous work investigated how contention could be addressed via scheduling. A contention-aware scheduler separates competing threads onto separate memory hierarchy domains to eliminate resource sharing and, as a consequence, mitigate contention. However, all previous work on contention-aware scheduling assumed that the underlying system is UMA (uniform memory access latencies, single memory controller). Modern multicore systems, however, are NUMA, which means that they feature non-uniform memory access latencies and multiple memory controllers. We discovered that contention management is a lot more difficult on NUMA systems, because the scheduler must not only consider the placement of threads, but also the placement of their memory. This is mostly required to eliminate contention for memory controllers contrary to the popular belief that remote access latency is the dominant concern. In this work we quantify the effects on performance imposed by resource contention and remote access latency. This analysis inspires the design of a contention-aware scheduling algorithm for NUMA systems. This algorithm significantly outperforms a NUMA-unaware algorithm proposed before as well as the default Linux scheduler. We also investigate memory migration strategies, which are the necessary part of the NUMA contention-aware scheduling algorithm. Finally, we propose and evaluate a new contention management algorithm that is priority-aware.
Subfields: Computer Networks and Communications, Hardware and Architecture
View Paper: https://www.semanticscholar.org/paper/dfa82000687105f78febc701ba55e482fa2d2bdd
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Networks and Communications
- Christos Kozyrakis
- David Lo
- Liqun Cheng
- Rama K. Govindaraju
- John Wilkes
- Alexandra Fedorova
- Parthasarathy Ranganathan
- Eric Tune
In Hardware and Architecture
- Liqun Cheng
- David Lo
- Rama K. Govindaraju
- Parthasarathy Ranganathan
- Christos Kozyrakis
- John Wilkes
- Alexandra Fedorova
- Christina Delimitrou
Citations in the corpus
- Heracles: Improving resource efficiency at scale (2015, background): “In multi-socket servers, one can isolate workloads across NUMA channels [9, 73], but this approach constrains DRAM capacity allocation and address interleaving.”
- Traffic management: a holistic approach to memory placement on NUMA systems (2013, methodology): “The most comprehensive work to date on NUMA-aware contention management is the DINO scheduler [5], which spreads memory intensive threads across memory domains and accordingly migrates the corresponding memory pages.”
- CPI2: CPU performance isolation for shared compute clusters (2013, methodology): “This could certainly be done: for example, Zhang et al. [40] used memory reference counts to approximate memory bandwidth consumption on SMP machines; West et al. [38] used cache miss and reference counts to estimate cache occupancy of competing threads on multicore machines; VM3 [20] pro.led a”