Snapshot-Free, Transparent, and Robust Memory Reclamation for Lock-Free Data Structures
May 20, 2019 Β· Declared Dead Β· π ACM-SIGPLAN Symposium on Programming Language Design and Implementation
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Ruslan Nikolaev, Binoy Ravindran
arXiv ID
1905.07903
Category
cs.DC: Distributed Computing
Citations
18
Venue
ACM-SIGPLAN Symposium on Programming Language Design and Implementation
Last Checked
3 months ago
Abstract
We present a family of safe memory reclamation schemes, Hyaline, which are fast, scalable, and transparent to the underlying lock-free data structures. Hyaline is based on reference counting - considered impractical for memory reclamation in the past due to high overheads. Hyaline uses reference counters only during reclamation, but not while accessing individual objects, which reduces overheads for object accesses. Since with reference counters, an arbitrary thread ends up freeing memory, Hyaline's reclamation workload is (almost) balanced across all threads, unlike most prior reclamation schemes such as epoch-based reclamation (EBR) or hazard pointers (HP). Hyaline often yields (excellent) EBR-grade performance with (good) HP-grade memory efficiency, which is a challenging tradeoff with all existing schemes. Hyaline schemes offer: (i) high performance; (ii) good memory efficiency; (iii) robustness: bounding memory usage even in the presence of stalled threads, a well-known problem with EBR; (iv) transparency: supporting virtually unbounded number of threads (or concurrent entities) that can be created and deleted dynamically, and effortlessly join existent workload; (v) autonomy: avoiding special OS mechanisms and being non-intrusive to runtime or compiler environments; (vi) simplicity: enabling easy integration into unmanaged C/C++ code; and (vii) generality: supporting many data structures. All existing schemes lack one or more properties. We have implemented and tested Hyaline on x86(-64), ARM32/64, PowerPC, and MIPS. The general approach requires LL/SC or double-width CAS, while a specialized version also works with single-width CAS. Our evaluation reveals that Hyaline's throughput is very high - it steadily outperforms EBR by 10% in one test and yields 2x gains in oversubscribed scenarios. Hyaline's superior memory efficiency is especially evident in read-dominated workloads
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
π Similar Papers
In the same crypt β Distributed Computing
R.I.P.
π»
Ghosted
R.I.P.
π»
Ghosted
Reproducing GW150914: the first observation of gravitational waves from a binary black hole merger
R.I.P.
π»
Ghosted
MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
R.I.P.
π»
Ghosted
Adaptive Federated Learning in Resource Constrained Edge Computing Systems
R.I.P.
π»
Ghosted
Edge Intelligence: Paving the Last Mile of Artificial Intelligence with Edge Computing
R.I.P.
π»
Ghosted
iFogSim: A Toolkit for Modeling and Simulation of Resource Management Techniques in Internet of Things, Edge and Fog Computing Environments
Died the same way β π» Ghosted
R.I.P.
π»
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
π»
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
π»
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
π»
Ghosted