A type system for finding upper resource bounds of multi-threaded programs with nested transactions (original) (raw)

Safe Typing for Transactional vs. Lock-Based Concurrency in Multi-threaded Java

2010

Many concurrency models have been developed for high-level programming languages such as Java. A trend here is towards more flexible concurrency control protocols, going beyond the original Java multithreading treatment based on lexically-scoped concurrency control mechanism. Two proposals supporting flexible, non-lexical concurrency control are the lockhandling via the Lock-classes in Java 5 and Transactional Featherweight Java (TFJ), an extension of Featherweight Java by transactions.

A dependently typed framework for static analysis of program execution costs

2006

This paper considers the use of dependent types to capture information about dynamic resource usage in a static type system. Dependent types allow us to give (explicit) proofs of properties with a program; we present a dependently typed core language TT, and define a framework within this language for representing size metrics and their properties. We give several examples of size bounded programs within this framework and show that we can construct proofs of their size bounds within TT.

A new approach to the semantics of Multithreaded Java

Java has integrated multithreading to a far greater extent than most programming languages. It is also one of the only languages that specifies and requires safety guarantees for improperly synchronized programs. It turns out that understanding these issues is far more subtle and difficult than was previously thought. The existing specification makes guarantees that prohibit standard and proposed compiler optimizations; it also omits guarantees that are necessary for safe execution of much existing code. Some guarantees that are made (e.g., type safety) raise tricky implementation issues when running unsynchronized code on SMPs with weak memory models.

Transactional execution of Java programs

2005

Parallel programming is difficult due to the complexity of dealing with conventional lock-based synchronization. To simplify parallel programming, there have been a number of proposals to support transactions directly in hardware and eliminate locks completely. Although hardware support for transactions has the potential to completely change the way parallel programs are written, initially transactions will be used to execute existing parallel programs. In this paper we investigate the implications of using transactions to execute existing parallel Java programs. Our results show that transactions can be used to support all aspects of Java multithreaded programs. Moreover, the conversion of a lock-based application into transactions is largely straightforward. The performance that these converted applications achieve is equal to or sometimes better than the original lock-based implementation.

A Uniform Transactional Execution Environment for Java

Lecture Notes in Computer Science, 2008

Transactional memory (TM) has recently emerged as an effective tool for extracting fine-grain parallelism from declarative critical sections. In order to make STM systems practical, significant effort has been made to integrate transactions into existing programming languages. Unfortunately, existing approaches fail to provide a simple implementation that permits lock-based and transaction-based abstractions to coexist seamlessly. Because of the fundamental semantic differences between locks and transactions, legacy applications or libraries written using locks can not be transparently used within atomic regions. To address these shortcomings, we implement a uniform transactional execution environment for Java programs in which transactions can be integrated with more traditional concurrency control constructs. Programmers can run arbitrary programs that utilize traditional mutual-exclusion-based programming techniques, execute new programs written with explicit transactional constructs, and freely combine abstractions that use both coding styles.

A type-based algorithm for the control-flow analysis of higher-order concurrent programs

Lecture Notes in Computer Science, 1997

We address, in a type-based framework, the problem of control-flow analysis for concurrent and functional languages. We present an efficient algorithm that propagates automatically types, communication effects and call graphs. The algorithm comes with a logical characterization that consists of a type proof system. The latter operates on a Concurrent ML core-syntax: a strongly typed, polymorphic kernel that supports higher-order functions and concurrency primitives. Effects are represented as algebraic terms that record communication effects resulting from channel creation, sending and receiving. Call graphs record function calls and are captured by a term algebra that is close to usual process algebras. Types are annotated with effects and call graphs. For the sake of flexibility, a subtyping relation is considered on the type algebra. We present the language syntax together with its static semantics that consists of the typing rules and an inference algorithm. The latter is proved to be consistent and complete with respect to the typing rules.

Bounded Session Types for Object Oriented Languages

Lecture Notes in Computer Science, 2007

A session takes place between two parties; after establishing a connection, each party interleaves local computations and communications (sending or receiving) with the other. Session types characterise such sessions in terms of the types of values communicated and the shape of protocols, and have been developed for the π-calculus, CORBA interfaces, and functional languages. We study the incorporation of session types into object-oriented languages through MOOSE, a multi-threaded language with session types, thread spawning, iterative and higher-order sessions. Our design aims to consistently integrate the objectoriented programming style and sessions, and to be able to treat various case studies from the literature. We describe the design of MOOSE, its syntax, operational semantics and type system, and develop a type inference system. After proving subject reduction, we establish the progress property: once a communication has been established, well-typed programs will never starve at communication points.

A type and effect system for deterministic parallel Java

Proceeding of the 24th ACM SIGPLAN conference on Object oriented programming systems languages and applications - OOPSLA 09, 2009

Today's shared-memory parallel programming models are complex and error-prone. While many parallel programs are intended to be deterministic, unanticipated thread interleavings can lead to subtle bugs and nondeterministic semantics. In this paper, we demonstrate that a practical type and effect system can simplify parallel programming by guaranteeing deterministic semantics with modular, compile-time type checking even in a rich, concurrent object-oriented language such as Java. We describe an object-oriented type and effect system that provides several new capabilities over previous systems for expressing deterministic parallel algorithms. We also describe a language called Deterministic Parallel Java (DPJ) that incorporates the new type system features, and we show that a core subset of DPJ is sound. We describe an experimental validation showing that DPJ can express a wide range of realistic parallel programs; that the new type system features are useful for such programs; and that the parallel programs exhibit good performance gains (coming close to or beating equivalent, nondeterministic multithreaded programs where those are available).

Type-aware transactions for faster concurrent code

Proceedings of the Eleventh European Conference on Computer Systems - EuroSys '16, 2016

It is often possible to improve a concurrent system's performance by leveraging the semantics of its datatypes. We build a new software transactional memory (STM) around this observation. A conventional STM tracks read-and writesets of memory words; even simple operations can generate large sets. Our STM, which we call STO, tracks abstract operations on transactional datatypes instead. Parts of the transactional commit protocol are delegated to these datatypes' implementations, which can use datatype semantics, and new commit protocol features, to reduce bookkeeping, limit false conflicts, and implement efficient concurrency control. We test these ideas on the STAMP benchmark suite for STM applications and on our own prior work, the Silo high-performance in-memory database, observing large performance improvements in both systems.