BayesPerf: minimizing performance monitoring errors using Bayesian statistics
Subho S. Banerjee, Saurabh Jha, Zbigniew Kalbarczyk, Ravishankar K. Iyer
Abstract
Hardware performance counters (HPCs) that measure low-level architectural and microarchitectural events provide dynamic contextual information about the state of the system. However, HPC measurements are error-prone due to non determinism (e.g., undercounting due to event multiplexing, or OS interrupt-handling behaviors). In this paper, we present BayesPerf, a system for quantifying uncertainty in HPC measurements by using a domain-driven Bayesian model that captures microarchitectural relationships between HPCs to jointly infer their values as probability distributions. We provide the design and implementation of an accelerator that allows for low-latency and low-power inference of the BayesPerf model for x86 and ppc64 CPUs. BayesPerf reduces the average error in HPC measurements from 40.1% to 7.6% when events are being multiplexed. The value of BayesPerf in real-time decision-making is illustrated with a simple example of scheduling of PCIe transfers.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
6 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Tintin: A Unified Hardware Performance Profiling Infrastructure to Uncover and Manage Uncertainty cites FIRM: An Intelligent Fine-grained Resource Management Framework for SLO-Oriented Microservices
- Tintin: A Unified Hardware Performance Profiling Infrastructure to Uncover and Manage Uncertainty cites TEA: Time-Proportional Event Analysis
- Tintin: A Unified Hardware Performance Profiling Infrastructure to Uncover and Manage Uncertainty cites TIP: Time-Proportional Instruction Profiling
- Tintin: A Unified Hardware Performance Profiling Infrastructure to Uncover and Manage Uncertainty cites SoK: The Challenges, Pitfalls, and Perils of Using Hardware Performance Counters for Security
- TEA: Time-Proportional Event Analysis cites TIP: Time-Proportional Instruction Profiling
- Chips Need DIP: Time-Proportional Per-Instruction Cycle Stacks at Dispatch cites Tintin: A Unified Hardware Performance Profiling Infrastructure to Uncover and Manage Uncertainty
- Chips Need DIP: Time-Proportional Per-Instruction Cycle Stacks at Dispatch cites TEA: Time-Proportional Event Analysis
- Chips Need DIP: Time-Proportional Per-Instruction Cycle Stacks at Dispatch cites TIP: Time-Proportional Instruction Profiling
Cited by top-tier papers6
- TIP: Time-Proportional Instruction ProfilingBjörn Gottschall, Lieven Eeckhout, Magnus JahreMICRO 2021 · 16 citations
- TEA: Time-Proportional Event AnalysisBjörn Gottschall, Lieven Eeckhout, Magnus JahreISCA 2023 · 11 citations
- Tintin: A Unified Hardware Performance Profiling Infrastructure to Uncover and Manage UncertaintyAo Li, Marion Sudvarg, Zihan Li, Sanjoy K. Baruah et al.OSDI 2025 · 2 citations
- Chips Need DIP: Time-Proportional Per-Instruction Cycle Stacks at DispatchSilvio Heverton Campelo de Santana, Joseph Rogers, Lieven Eeckhout, Magnus JahreASPLOS 2026 · 1 citation
- CounterPoint: Using Hardware Event Counters to Refute and Refine Microarchitectural AssumptionsNick Lindsay, Caroline Trippel, Anurag Khandelwal, Abhishek BhattacharjeeASPLOS 2026
Builds on3
- FIRM: An Intelligent Fine-grained Resource Management Framework for SLO-Oriented MicroservicesHaoran Qiu, Subho S. Banerjee, Saurabh Jha, Zbigniew T. Kalbarczyk et al.OSDI 2020 · 350 citations
- SoK: The Challenges, Pitfalls, and Perils of Using Hardware Performance Counters for SecuritySanjeev Das, Jan Werner, Manos Antonakakis, Michalis Polychronakis et al.S&P 2019 · 163 citations
- Inductive-bias-driven Reinforcement Learning For Efficient Schedules in Heterogeneous ClustersSubho S. Banerjee, Saurabh Jha, Zbigniew Kalbarczyk, Ravishankar K. IyerICML 2020 · 15 citations
Related papers
- On the Feasibility of Malware Unpacking via Hardware-assisted Loop ProfilingBinlin Cheng, Erika A. Leal, Haotian Zhang, Jiang MingUSENIX Security 2023
- The Laplace Microarchitecture for Tracking Data Uncertainty and Its Implementation in a RISC-V ProcessorVasileios Tsoutsouras, Orestis Kaparounakis, Bilgesu Arif Bilgin, Chatura Samarakoon et al.MICRO 2021 · 5 citations
- Data-Driven Structured Thermal Modeling for COTS Multi-core ProcessorsSeyedmehdi Hosseinimotlagh, Daniel Enright, Christian R. Shelton, Hyoseung KimRTSS 2021 · 4 citations
- Fast-BCNN: Massive Neuron Skipping in Bayesian Convolutional Neural NetworksQiyu Wan, Xin FuMICRO 2020 · 26 citations
- Enabling fast uncertainty estimation: accelerating bayesian transformers via algorithmic and hardware optimizationsHongxiang Fan, Martin Ferianc, Wayne LukDAC 2022 · 7 citations
