2019 · 59 citations · 45 references
ComputingComputer EngineeringComputer ScienceCloud Performance OptimizationParallel AlgorithmsParallel ComputingCpu SchedulingDistributed ComputingLatency OptimizationLoad BalancingMulticore Processor DesignScalability ArchitectureAdaptive Load BalancingAdaptive ScalingDynamic Load BalancingDynamic ScalingLoad Balancing AlgorithmsLoad Balancing StrategiesServer Load BalancingTail Latency
While the current literature typically focuses on load-balancing among multiple servers, in this paper, we demonstrate the importance of load-balancing within a single machine (potentially with hundreds of CPU cores). In this context, we propose a new load-balancing technique (RSS++) that dynamically modifies the receive side scaling (RSS) indirection table to spread the load across the CPU cores in a more optimal way. RSS++ incurs up to 14x lower 95th percentile tail latency and orders of magnitude fewer packet drops compared to RSS under high CPU utilization. RSS++ allows higher CPU utilization and dynamic scaling of the number of allocated CPU cores to accommodate the input load while avoiding the typical 25% over-provisioning.
45
Eddie Kohler, Robert Morris, Benjie Chen et al. · ACM Transactions on Computer Systems · 2000 · 2.4K citations · Full text
Consistent hashing and random trees
David R. Karger, Eric Lehman, Tom Leighton et al. · 1997 · 2K citations
Rasmus Pagh, Flemming Friche Rodler · Journal of Algorithms · 2004 · 1.1K citations
Inside the Social Network's (Datacenter) Network
Arjun Roy, Hongyi Zeng, Jasmeet Bagga et al. · 2015 · 829 citations · Full text
Eli Cortez, Anand Bonde, Alexandre Muzio et al. · 2017 · 592 citations