Constant-Time Root Scanning for Deterministic Garbage Collection

Siebert, Fridtjof

doi:10.1007/3-540-45306-7_21

Cited by 24 publications

(27 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another good example of incremental root scanning is Siebert's approach [93]. As it is a part of a collector which can provide a certain degree of real-time guarantees, it will be introduced along with the collector in the next section.…”

Section: Incremental Root Scanningmentioning

confidence: 99%

See 1 more Smart Citation

Garbage Collection for Flexible Hard Real-Time Systems

Chang

Wellings

2010

IEEE Trans. Comput.

View full text Add to dashboard Cite

Garbage collection is key to improving the productivity and code quality in virtually every computing system with dynamic memory requirements. It dramatically reduces the software complexity and consequently allows the programmers to concentrate on the real problems rather than struggling with memory management issues, which finally results in higher productivity. The code quality is also improved because the two most common memory related errors -dangling pointers and memory leaks -can be prevented (assuming that the garbage collectors are implemented correctly and the user programs are not malicious).However, real-time systems, particularly those with hard real-time requirements, always choose not to use garbage collection (or even dynamic memory) in order to avoid the unpredictable executions of garbage collectors as well as the chances of being blocked due to the lack of free memory. Much effort has been expended trying to build predictable garbage collectors which can provide both temporal and spatial guarantees. Unfortunately, most existing work leads to systems that cannot easily achieve a balance between temporal and spatial performance, although their worst-case behaviours in both dimensions can be predicted. Moreover, the real-time systems targeted by the existing real-time garbage collectors are not the state-ofthe-art ones. The scheduling of those garbage collectors has not been integrated into the modern real-time scheduling frameworks, which makes the benefits provided by those systems very difficult to obtain. This thesis argues that the aforementioned issues should be tackled by introducing new design criteria for real-time garbage collectors. The existing criteria are not enough to reflect the unique requirements of real-time systems. A new performance indicator is proposed to describe the capability of a real-time garbage collector to achieve better balance between temporal and spatial performance. Moreover, it is argued that new real-time garbage collectors should be integrated with the real-time task model and more advanced scheduling techniques should be adopted as well.A hybrid garbage collection algorithm, which combines both reference counting and mark-sweep, is designed according to the new design criteria. On the one hand, the reference counting component helps reduce the overall memory requirements. On the other hand, the mark-sweep component periodically identifies cyclic garbage, which cannot be found by the reference counting component. Both parts of the collector are executed by segregated tasks, which are released periodically as the hard real-time user tasks. In order to improve the performance of soft-or non-realtime tasks in our system while still guaranteeing the hard real-time requirements, the dual-priority scheduling algorithm is used for all the tasks including the GC tasks. A multi-heap approach is also proposed to bound the impacts of the softor non-real-time tasks on the overall memory usage as well as the executions of the GC tasks. More importantly, static ana...

show abstract

Section: Incremental Root Scanningmentioning

confidence: 99%

“…Siebert published several papers and a thesis to present his unique Dijkstra style write barrier based incremental mark-sweep garbage collector which is claimed to be real-time [90,91,89,92,93,95,94]. One of his two major contributions is the introduction of a new way to eliminate fragmentation.…”

Section: Siebert's Algorithmmentioning

confidence: 99%

Garbage Collection for Flexible Hard Real-Time Systems

Chang

Wellings

2010

IEEE Trans. Comput.

View full text Add to dashboard Cite

show abstract

“…Memory management mechanisms, such as dynamic memory allocators [26] and garbage collectors [48], can introduce additional unpredictable overheads. On top of this, address space layout randomization and the physical page allocation mechanism change the logical and physical memory layout of the application every time it is executed, potentially affecting the number of conflict misses in the CPU caches and branch mispredictions [17,38].…”

Section: Introductionmentioning

confidence: 99%

Minimizing the cost of iterative compilation with active learning

William

Petoumenos

Wang

et al. 2017

2017 IEEE/ACM International Symposium on Code Generation and Optimization (CGO)

View full text Add to dashboard Cite

Since performance is not portable between platforms, engineers must fine-tune heuristics for each processor in turn. This is such a laborious task that high-profile compilers, supporting many architectures, cannot keep up with hardware innovation and are actually out-of-date. Iterative compilation driven by machine learning has been shown to be efficient at generating portable optimization models automatically. However, good quality models require costly, repetitive, and extensive training which greatly hinders the wide adoption of this powerful technique.In this work, we show that much of this cost is spent collecting training data, runtime measurements for different optimization decisions, which contribute little to the final heuristic. Current implementations evaluate randomly chosen, often redundant, training examples a pre-configured, almost always excessive, number of times -a large source of wasted effort. Our approach optimizes not only the selection of training examples but also the number of samples per example, independently. To evaluate, we construct 11 high-quality models which use a combination of optimization settings to predict the runtime of benchmarks from the SPAPT suite. Our novel, broadly applicable, methodology is able to reduce the training overhead by up to 26x compared to an approach with a fixed number of sample runs, transforming what is potentially months of work into days.

show abstract

“…Techniques for accurate stack scanning in uncooperative environments have been previously described in detail in [13,8]. Popular techniques for conservative garbage collection include the Boehm-Weiser collector [2] and various incarnations of mostly-copying collectors [14,15,16].…”

Section: Related Workmentioning

confidence: 99%

“…JamaicaVM uses explicit pointer stacks [13], but they differ from our implementation. First, objects referenced from the stack cannot move (in Ovm they can).…”

Section: Related Workmentioning

confidence: 99%

Accurate Garbage Collection in Uncooperative Environments with Lazy Pointer Stacks

Baker

Cunei

Pizlo

et al.

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Implementing a new programming language by the means of a translator to an existing language is attractive as it provides portability over all platforms supported by the host language and reduces the development time as many low-level tasks can be delegated to the host compiler. The C and C++ programming languages are popular choices for many language implementations due to the availability of efficient compilers on many platforms, and good portability. For garbage-collected languages, however, they are not a perfect match as they provide no support for accurately discovering pointers to heap-allocated data. We evaluate the published techniques, and propose a new mechanism, lazy pointer stacks, for performing accurate garbage collection in such uncooperative environments. We implemented the new technique in the Ovm Java virtual machine with our own Java-to-C++ compiler and GCC as a back-end, and found that our technique outperforms existing approaches.

show abstract

Constant-Time Root Scanning for Deterministic Garbage Collection

Cited by 24 publications

References 21 publications

Garbage Collection for Flexible Hard Real-Time Systems

Garbage Collection for Flexible Hard Real-Time Systems

Minimizing the cost of iterative compilation with active learning

Accurate Garbage Collection in Uncooperative Environments with Lazy Pointer Stacks

Contact Info

Product

Resources

About