Declarative aspects of memory management in the concurrent collections parallel programming model

Abstract

Concurrent Collections (CnC) [8] is a declarative parallel language that allows the application developer to express their parallel application as a collection of high-level computations called steps that communicate via single-assignment data structures called items.A CnC program is specified in two levels. At the bottom level, an existing imperative language implements the computations within the individual computation steps. At the top level, CnC describes the relationships (ordering constraints) among the steps. The memory management mechanism of the existing imperative language manages data whose lifetime is within a computation step. A key limitation in the use of CnC for long-running programs is the lack of memory management and garbage collection for data items with lifetimes that are longer than a single computation step. Although the goal here is the same as that of classical garbage collection, the nature of problem and therefore nature of the solution is distinct. The focus of this paper is the memory management problem for these data items in CnC.We introduce a new declarative slicing annotation for CnC that can be transformed into a reference counting procedure for memory management. Preliminary experimental results obtained from a Cholesky example show that our memory management approach can result in space reductions for CnC data items of up to 28× relative to the baseline case of standard CnC without memory management.

SCnC: Efficient Unification of Streaming with Dynamic Task Parallelism

“…The performance analysis focuses on the variation of throughput between implementations and not on absolute timing of applications, as the current CnC implementation has memory collection problems that lead to OutOfMemory exceptions for long running programs such as streaming ones [7]. The size of the problem used for CnC is the maximum size for which the garbage collection does not go over 10% of application runtime.…”

Extreme-scale computer architecture: Energy efficiency from the ground up

“…One such model is Hierarchical Tiled Arrays (HTA) [23], which allows the computation to be expressed in recursive blocks or tiles. Another possible model is Concurrent Collections [24], which expresses the program in a dataflow-like manner. These are high-level models, and the compiler still has to translate them into efficient machine code.…”

Identifying a Unifying Mechanism for the Implementation of Concurrency Abstractions on Multi-language Virtual Machines

“…Today's high-level language virtual machines (VMs) do not provide intrinsic support for more than one specific approach [17]. While some approaches like Fork/Join [14], Concurrent Collections [3] or PLINQ can be implemented as libraries without losing any semantics or performance, approaches like the actor model are typically implemented as an approximation losing for instance the engineering benefits of encapsulation [11].…”