Generational Garbage Collection Research Articles

Optimizations of Generational Garbage Collections Based on Continuous Damage Process

In the computer science community, garbage collection is a dynamic storage management technology to ensure the reliability of computer systems. In this paper, we consider two garbage collection policies to meet the goal of time consumption for a generational garbage collector when increase in objects might be unclear at discrete times for the high frequency of computer processes. That is, (a) tenuring collection is triggered at the [Formula: see text]th minor collection preventively or at a threshold amount [Formula: see text] of surviving objects correctively, and (b) major collection is made at discrete times [Formula: see text] for a given [Formula: see text] or at the [Formula: see text]th collection including minor and tenuring collections. Using the damage process and renewal theory, the expected cost rates are obtained, and their optimal policies for tenuring and major collection are discussed analytically and computed numerically.

Optimizations of discrete garbage collections in computer reliability systems

In computer science community, garbage collection is a dynamic storage management technology to ensure the reliability of computer systems. In this paper, we consider two discrete garbage collection policies to meet the goal of time consumption for a generational garbage collector. That is, garbage collections occur at a nonhomogeneous Poisson process, (a) tenuring collection is triggered at the Nth minor collection preventively or at a threshold amount [Formula: see text] of surviving objects correctively, whichever takes place first, and (b), tenuring collection is triggered at the first collection when the amount of surviving objects has exceeded a threshold level [Formula: see text] and major collection is triggered at discrete times kT for a given T. Using the damage process and renewal theory, the expected cost rates are obtained, and their optimal policies for tenuring and major collections are discussed analytically and computed numerically.

Integrating region memory management and tag-free generational garbage collection

Abstract We present a region-based memory management scheme with support for generational garbage collection. The scheme features a compile-time region inference algorithm, which associates values with logical regions, and builds on a region type system that deploys region types at runtime to avoid the overhead of write barriers and to support partly tag-free garbage collection. The scheme is implemented in the MLKit Standard ML compiler, which generates native x64 machine code. Besides demonstrating a number of important formal properties of the scheme, we measure the scheme’s characteristics, for a number of benchmarks, and compare the performance of the generated executables with the performance of executables generated with the MLton state-of-the-art Standard ML compiler and configurations of the MLKit with and without region inference and generational garbage collection enabled. Although region inference often serves the purpose of generations, combining region inference with generational garbage collection is shown often to be superior to combining region inference with non-generational collection despite the overhead introduced by a larger amount of memory waste, due to region fragmentation.

Certifying graph-manipulating C programs via localizations within data structures

We develop powerful and general techniques to mechanically verify realistic programs that manipulate heap-represented graphs. These graphs can exhibit well-known organization principles, such as being a directed acyclic graph or a disjoint-forest; alternatively, these graphs can be totally unstructured. The common thread for such structures is that they exhibit deep intrinsic sharing and can be expressed using the language of graph theory. We construct a modular and general setup for reasoning about abstract mathematical graphs and use separation logic to define how such abstract graphs are represented concretely in the heap. We develop a Localize rule that enables modular reasoning about such programs, and show how this rule can support existential quantifiers in postconditions and smoothly handle modified program variables. We demonstrate the generality and power of our techniques by integrating them into the Verified Software Toolchain and certifying the correctness of seven graph-manipulating programs written in CompCert C, including a 400-line generational garbage collector for the CertiCoq project. While doing so, we identify two places where the semantics of C is too weak to define generational garbage collectors of the sort used in the OCaml runtime. Our proofs are entirely machine-checked in Coq.

A Verified Generational Garbage Collector for CakeML

This paper presents the verification of a generational copying garbage collector for the CakeML runtime system. The proof is split into an algorithm proof and an implementation proof. The algorithm proof follows the structure of the informal intuition for the generational collector’s correctness, namely, a partial collection cycle in a generational collector is the same as running a full collection on part of the heap, if one views pointers to old data as non-pointers. We present a pragmatic way of dealing with ML-style mutable state, such as references and arrays, in the proofs. The development has been fully integrated into the in-logic bootstrapped CakeML compiler, which now includes command-line arguments that allow configuration of the generational collector. All proofs were carried out in the HOL4 theorem prover.

Dynamic atomicity: optimizing swift memory management

Swift is a modern multi-paradigm programming language with an extensive developer community and open source ecosystem. Swift 3's memory management strategy is based on Automatic Reference Counting (ARC) augmented with unsafe APIs for manually-managed memory. We have seen ARC consume as much as 80% of program execution time. A significant portion of ARC's direct performance cost can be attributed to its use of atomic machine instructions to protect reference count updates from data races. Consequently, we have designed and implemented dynamic atomicity, an optimization which safely replaces atomic reference-counting operations with nonatomic ones where feasible. The optimization introduces a store barrier to detect possibly intra-thread references, compiler-generated recursive reference-tracers to find all affected objects, and a bit of state in each reference count to encode its atomicity requirements. Using a suite of 171 microbenchmarks, 9 programs from the Computer Language Benchmarks Game, and the Richards benchmark, we performed a limit study by unsafely making all reference counting operations nonatomic. We measured potential speedups of up to 220% on the microbenchmarks, 120% on the Benchmarks Game and 70% on Richards. By automatically reducing ARC overhead, our optimization both improves Swift 3's performance and reduces the temptation for performance-oriented programmers to resort to unsafe manual memory management. Furthermore, the machinery implemented for dynamic atomicity could also be employed to obtain cheaper thread-safe Swift data structures, or to augment ARC with optional cycle detection or a backup tracing garbage collector.

Automating Live Update for Generic Server Programs

The pressing demand to deploy software updates without stopping running programs has fostered much research on live update systems in the past decades. Prior solutions, however, either make strong assumptions on the nature of the update or require extensive and error-prone manual effort, factors which discourage the adoption of live update. This paper presents <i>Mutable Checkpoint-Restart</i> ( <i>MCR</i> ), a new live update solution for generic (multiprocess and multithreaded) server programs written in C. Compared to prior solutions, MCR can support arbitrary software updates and automate most of the common live update operations. The key idea is to allow the running version to safely reach a quiescent state and then allow the new version to restart as similarly to a fresh program initialization as possible, relying on existing code paths to automatically restore the old program threads and reinitialize a relevant portion of the program data structures. To transfer the remaining data structures, MCR relies on a combination of precise and conservative garbage collection techniques to trace all the global pointers and apply the required state transformations on the fly. Experimental results on popular server programs ( <i>Apache httpd</i> , <i>nginx</i> , <i>OpenSSH</i> and <i>vsftpd</i> ) confirm that our techniques can effectively automate problems previously deemed difficult at the cost of negligible performance overhead (2 percent on average) and moderate memory overhead (3.9 <inline-formula><tex-math notation="LaTeX">$\times$ </tex-math></inline-formula> on average, without optimizations).

A Novel Design of a Parallel Machine Learnt Generational Garbage Collector

The Generational Garbage collection involves organizing the heap into different divisions of memory space in-order to filter long-lived objects from short-lived objects through moving the surviving object of each generation’s GC cycle to another memory space, updating its age and reclaiming space from the dead ones. The problem in this method is that, the longer an object is alive during its initial generations, the longer the garbage collector will have to deal with it by checking for its reachability from the root and promoting it to other space divisions, where as the ultimate goal of the GC is to reclaim memory from unreachable objects at a minimal time possible. This paper is a proposal of a method where the lifetime of every object getting into the heap will be predicted and will be placed in heap accordingly for the garbage collector to deal more with reclaiming space from dead object and less in promoting the live ones to the higher level.

Taking off the gloves with reference counting Immix

Despite some clear advantages and recent advances, reference counting remains a poor cousin to high-performance tracing garbage collectors. The advantages of reference counting include a) immediacy of reclamation, b) incrementality, and c) local scope of its operations. After decades of languishing with hopelessly bad performance, recent work narrowed the gap between reference counting and the fastest tracing collectors to within 10%. Though a major advance, this gap remains a substantial barrier to adoption in performance-conscious application domains. Our work identifies heap organization as the principal source of the remaining performance gap. We present the design, implementation, and analysis of a new collector, RC Immix, that replaces reference counting's traditional free-list heap organization with the line and block heap structure introduced by the Immix collector. The key innovations of RC Immix are 1) to combine traditional reference counts with per-line live object counts to identify reusable memory and 2) to eliminate fragmentation by integrating copying with reference counting of new objects and with backup tracing cycle collection. In RC Immix, reference counting offers efficient collection and the line and block heap organization delivers excellent mutator locality and efficient allocation. With these advances, RC Immix closes the 10% performance gap, matching the performance of a highly tuned production generational collector. By removing the performance barrier, this work transforms reference counting into a serious alternative for meeting high performance objectives for garbage collected languages.

Efficient context sensitivity for dynamic analyses via calling context uptrees and customized memory management

State-of-the-art dynamic bug detectors such as data race and memory leak detectors report program locations that are likely causes of bugs. However, programmers need more than static program locations to understand the behavior of increasingly complex and concurrent software. Dynamic calling context provides additional information, but it is expensive to record calling context frequently, e.g., at every read and write. Context-sensitive dynamic analyses can build and maintain a calling context tree (CCT) to track calling context--but in order to reuse existing nodes, CCT-based approaches require an expensive lookup. This paper introduces a new approach for context sensitivity that avoids this expensive lookup. The approach uses a new data structure called the calling context uptree (CCU) that adds low overhead by avoiding the lookup and instead allocating a new node for each context. A key contribution is that the approach can mitigate the costs of allocating many nodes by extending tracing garbage collection (GC): GC collects unused CCU nodes naturally and efficiently, and we extend GC to merge duplicate nodes lazily. We implement our CCU-based approach in a high-performance Java virtual machine and integrate it with a staleness-based memory leak detector and happens-before data race detector, so they can report context-sensitive program locations that cause bugs. We show that the CCU-based approach, in concert with an extended GC, provides a compelling alternative to CCT-based approaches for adding context sensitivity to dynamic analyses.

Tutorial overview

In spite of the high-level abstraction benefits of automatic tracing garbage collection, current prevailing sentiment within the safety certification community is that a simpler memory model is required for the most rigorous levels of software safety certification. Thus, the draft JSR-302 specification for safety critical Java relies on scope-based memory allocation rather than tracing garbage collection. The scoped memory model for JSR-302 is a simplification of the RTSJ model. JSR-302 enforces a strict hierarchy of scopes and distinguishes private scopes, which can be seen only by one thread, from mission scopes, which can be accessed by all the threads that comprise a mission, including threads running within inner-nested sub-missions. The hierarchical memory structure allows implementations to guarantee the absence of memory fragmentation for scope management, unlike the Real-Time Specification for Java from which the JSR-302 specification was derived. In the absence of block structure, it is more difficult in Java to safely manage references to stack-allocated objects than in Ada. While the simplified hierarchical management of scoped memory that is part of JSR-302 addresses memory fragmentation concerns, it does not guarantee the absence of dangling pointers. As with the Real-Time Specification for Java, JSR-302 requires a run-time check to enforce that no reference assignment creates a relationship whereby an outer-nested object is allowed to point to an inner-nested object. This rule assures the absence of dangling pointers, but it introduces a different problem: every assignment to a reference field must be accompanied by a run-time check to validate the appropriate scope nesting relationship. This run-time check will throw a run-time exception if the assignment is deemed inappropriate. The safety certification evidence for a given safety-critical Java program must therefore include an argument for every reference assignment that it will not cause the program to abort with a run-time exception. Furthermore, the certification evidence must prove that sufficient memory is available to reliably execute each safety-critical task in the system. This tutorial provides an overview of dynamic memory management in Safety Critical Java and describes two annotation systems that have been designed to support static (compile-time) enforcement of memory safety properties. The first annotation system is described in an appendix to the draft JSR-302 standard. This relatively simple annotation system, which is not considered normative, serves to demonstrate that memory safety can be statically proven without requiring extensive annotations throughout existing library code. The second annotation system is the system implemented in Perc Pico. This annotation system, which is much richer than the draft JSR-302 annotation, has been in experimental use for over five years. During that time, tens of thousands of lines of experimental application code have been developed, with the experience motivating a variety of refinements to the original design.

OPTIMAL TENURING AND MAJOR COLLECTION TIMES FOR A GENERATIONAL GARBAGE COLLECTOR

It is an important problem to determine the tenuring collection time or major collection time to meet the pause time goal for a generational garbage collector. From such a viewpoint, this paper proposes two stochastic models based on the working schemes of a generational garbage collector: Garbage collections occur at a nonhomogeneous Poisson process. Minor collections are made when the garbage collector begins to work, tenuring collection is made at a planned time T or at the first collection time when surviving objects have exceeded K for the first model. Major collection is made at time T or at the Nth collection for the second model. Using the techniques of cumulative processes and reliability theory, expected cost rates are obtained, and optimal policies of tenuring and major collection times which minimize them are discussed analytically and computed numerically.

Multitasking without compromise

The Multitasking Virtual Machine (called from now on simply MVM) is a modification of the Java™ virtual machine. It enables safe, secure, and scalable multitasking. Safety is achieved by strict isolation of applications from one another. Resource control mechanisms augment security by preventing some denial-ofservice attacks. Improved scalability results from an aggressive application of the main design principle of MVM: share as much of the runtime as possible among applications and replicate everything else. The system can be described as a 'no compromise' approach -- all the known APIs and mechanisms of the Java programming language are available to applications. MVM is implemented as a series of carefully tuned modifications to the Java HotSpot™ virtual machine, including the dynamic compiler. This paper presents the design of MVM, focusing on several novel and general techniques: an in-runtime design of lightweight isolation, an extension of a copying, generational garbage collector to provide best-effort management of a portion of the heap space, and a transparent and automated mechanism for safe execution of user-level native code. MVM demonstrates that multitasking in a safe language can be accomplished with a high degree of protection, without constraining the language, and with competitive performance characteristics.

Two Generational Garbage Collection Models with Major Collection Time

It is an important problem to determine major collection times to meet the pause time goal for a generational garbage collector. From such a viewpoint, this paper proposes two stochastic models based on working schemes of a generational garbage collector: Garbage collections occur in a nonhomogeneous Poisson process, tenuring collection is made at a threshold level K, and major collection is made at time T or at Nth collection including minor and tenuring collections for the first model and at time T or at Nth collection including tenuring collections for the second model. Using the techniques of cumulative processes and reliability theory, expected cost rates are obtained, and optimal policies of major collection times which minimize them are discussed analytically and computed numerically.

Nonblocking real-time garbage collection

A real-time garbage collector has to fulfill two basic properties: ensure that programs with bounded allocation rates do not run out of memory and provide short blocking times. Even for incremental garbage collectors, two major sources of blocking exist, namely, root scanning and heap compaction. Finding root nodes of an object graph is an integral part of tracing garbage collectors and cannot be circumvented. Heap compaction is necessary to avoid probably unbounded heap fragmentation, which in turn would lead to unacceptably high memory consumption. In this article, we propose solutions to both issues. Thread stacks are local to a thread, and root scanning, therefore, only needs to be atomic with respect to the thread whose stack is scanned. This fact can be utilized by either blocking only the thread whose stack is scanned, or by delegating the responsibility for root scanning to the application threads. The latter solution eliminates blocking due to root scanning completely. The impact of this solution on the execution time of a garbage collector is shown for two different variants of such a root scanning algorithm. During heap compaction, objects are copied. Copying is usually performed atomically to avoid interference with application threads, which could render the state of an object inconsistent. Copying of large objects and especially large arrays introduces long blocking times that are unacceptable for real-time systems. In this article, an interruptible copy unit is presented that implements nonblocking object copy. The unit can be interrupted after a single word move. We evaluate a real-time garbage collector that uses the proposed techniques on a Java processor. With this garbage collector, it is possible to run high-priority hard real-time tasks at 10 kHz parallel to the garbage collection task on a 100 MHz system.

Tracing garbage collection on highly parallel platforms

The pervasiveness of multiprocessor and multicore hardware and the rising level of available parallelism are radically changing the computing landscape. Can software deal with tomorrow's potential higher parallelism? In this paper we study this issue from the garbage collection perspective. In particular, we investigate the scalability of parallel heap tracing, which stands at the core of the garbage collection activity. Heap shapes can be sequential in nature, and prevent the collector from scaling the trace. We start by proposing the idealized trace utilization as a scalability measure for evaluating the scalability of a given heap shape. We then examine standard Java benchmarks and evaluate the existence of non-scalable object-graph shapes in their execution. Next, we propose and implement a prototype of garbage collection techniques that attempt to ameliorate the object-graph shape problem. Finally, we measure and report their efficacy.

Dynamic filtering

This paper introduces a new abstraction to accelerate the read-barriers and write-barriers used by language runtime systems. We exploit the fact that, dynamically, many barrier executions perform checks but no real work -- e.g., in generational garbage collection (GC), frequent checks are needed to detect the creation of inter-generational references, even though such references occur rarely in many workloads. We introduce a form of dynamic filtering that identifies redundant checks by (i) recording checks that have recently been executed, and (ii) detecting when a barrier is repeating one of these checks. We show how this technique can be applied to a variety of algorithms for GC, transactional memory, and language-based security. By supporting dynamic filtering in the instruction set, we show that the fast-paths of these barriers can be streamlined, reducing the impact on the quality of surrounding code. We show how we accelerate the barriers used for generational GC and transactional memory in the Bartok research compiler. With a 2048-entry filter, dynamic filtering eliminates almost all the overhead of the GC write-barriers. Dynamic filtering eliminates around half the overhead of STM over a non-synchronized baseline -- even when used with an STM that is already designed for low overhead, and which employs static analyses to avoid redundant operations.

Dynamic filtering

This paper introduces a new abstraction to accelerate the read-barriers and write-barriers used by language runtime systems. We exploit the fact that, dynamically, many barrier executions perform checks but no real work -- e.g., in generational garbage collection (GC), frequent checks are needed to detect the creation of inter-generational references, even though such references occur rarely in many workloads. We introduce a form of dynamic filtering that identifies redundant checks by (i) recording checks that have recently been executed, and (ii) detecting when a barrier is repeating one of these checks. We show how this technique can be applied to a variety of algorithms for GC, transactional memory, and language-based security. By supporting dynamic filtering in the instruction set, we show that the fast-paths of these barriers can be streamlined, reducing the impact on the quality of surrounding code. We show how we accelerate the barriers used for generational GC and transactional memory in the Bartok research compiler. With a 2048-entry filter, dynamic filtering eliminates almost all the overhead of the GC write-barriers. Dynamic filtering eliminates around half the overhead of STM over a non-synchronized baseline -- even when used with an STM that is already designed for low overhead, and which employs static analyses to avoid redundant operations.

Flexible reference-counting-based hardware acceleration for garbage collection

Languages featuring automatic memory management (garbage collection) are increasingly used to write all kinds of applications because they provide clear software engineering and security advantages. Unfortunately, garbage collection imposes a toll on performance and introduces pause times, making such languages less attractive for high-performance or real-time applications. Much progress has been made over the last five decades to reduce the overhead of garbage collection, but it remains significant. We propose a cooperative hardware-software technique to reduce the performance overhead of garbage collection. The key idea is to reduce the frequency of garbage collection by efficiently detecting and reusing dead memory space in hardware via hardware-implemented reference counting. Thus, even though software garbage collections are still eventually needed, they become much less frequent and have less impact on overall performance. Our technique is compatible with a variety of software garbage collection algorithms, does not break compatibility with existing software, and reduces garbage collection time by 31% on average on the Java DaCapo benchmarks running on the production build of the Jikes RVM, which uses a state-of-the-art generational garbage collector.

Immix

Programmers are increasingly choosing managed languages for modern applications, which tend to allocate many short-to-medium lived small objects. The garbage collector therefore directly determines program performance by making a classic space-time tradeoff that seeks to provide space efficiency, fast reclamation, and mutator performance. The three canonical tracing garbage collectors: semi-space, mark-sweep, and mark-compact each sacrifice one objective. This paper describes a collector family, called mark-region , and introduces opportunistic defragmentation, which mixes copying and marking in a single pass. Combining both, we implement immix , a novel high performance garbage collector that achieves all three performance objectives. The key insight is to allocate and reclaim memory in contiguous regions, at a coarse block grain when possible and otherwise in groups of finer grain lines . We show that immix outperforms existing canonical algorithms, improving total application performance by 7 to 25% on average across 20 benchmarks. As the mature space in a generational collector, immix matches or beats a highly tuned generational collector, e.g. it improves jbb2000 by 5%. These innovations and the identification of a new family of collectors open new opportunities for garbage collector design.