Memory management for parallel tasks in shared memory

2005 ◽  
pp. 165-178
Author(s):  
K. G. Langendoen ◽  
H. L. Muller ◽  
W. G. Vree
Keyword(s):  
Shared Memory ◽  
Memory Management ◽  
Parallel Tasks

PARALLEL BUDDY MEMORY MANAGEMENT

1992 ◽  
Vol 02 (04) ◽  
pp. 391-398 ◽  
Author(s):  
THEODORE JOHNSON ◽  
TIMOTHY A. DAVIS
Keyword(s):  
Shared Memory ◽  
Memory Management ◽  
Sparse Matrix ◽  
Parallel Programs ◽  
Memory Allocation ◽  
Multiprocessor Systems ◽  
Matrix Algorithm ◽  
Buddy System ◽  
Shared Memory Multiprocessor ◽  
Dynamic Storage

Shared memory multiprocessor systems need efficient dynamic storage allocators, both for system purposes and to support parallel programs. Memory managers are often based on the buddy system, which provides fast allocation and release. Previous parallel buddy memory managers made no attempt to coordinate the allocation, splitting and release of blocks, and as a result needlessly fragment memory. We present a fast and simple parallel buddy memory manager that is also as space efficient as a serial buddy memory manager. We test our algorithms using memory allocation/deallocation traces collected from a parallel sparse matrix algorithm.

scholarly journals L2 cache performance analysis for MPSoC by tag – comparision

2017 ◽  
Vol 7 (1.5) ◽  
pp. 285
Author(s):  
Jenitha A ◽  
Elumalai R
Keyword(s):  
Performance Analysis ◽  
Shared Memory ◽  
Memory Management ◽  
Memory Systems ◽  
Critical Problem ◽  
Energy Current ◽  
L2 Cache ◽  
Application Processor ◽  
Cache Miss ◽  
Embedded Applications

Memory systems in many applications are becoming increasingly large, contributing to many challenges in the memory management that has led to many method to manage memory. The tag comparison consumes large amount of cache energy. Current methods provide tag comparison cache or failure of the expected cache. Here is proposed an idea based on new call Comparing Tag stages, filter bloom is presented to improve the efficiency of the cache to predict failure and partial tag comparison for the cold line of verification and full comparison check for direct labels. Moreover, the administration of the cache that is filled with cache lines occurs when there is a cache miss. Today's embedded applications use MPSoC. The  MPSoC consists of the following ie more than one  processors, shared memory among the processors available and a global  off-chip memory. Planning of the activities of an integrated application processor and memory partition between processors are two main critical problem. Here, for an integrated application, both task scheduling and partitioning the integrated available L2 cache to reduce the runtime approach is used.

Современные интервальные подходы к вычислению решений непрерывных игр 

2018 ◽  
Author(s):  
B.J. Kubica
Keyword(s):  
Shared Memory ◽  
Branch And Bound ◽  
Memory Management ◽  
Nash Equilibria ◽  
Automatic Differentiation ◽  
Difficult Problem ◽  
Interval Methods ◽  
Interval Matrices ◽  
Management Techniques ◽  
Interval Branch And Bound

Computing Nash equilibria in continuous games is a difficult problem, but interval methods have already been applied to its solution quite successfully. The purpose of this paper is to briefly survey previous efforts and achievements of the author related to the topic, and to consider some advanced tools for accelerating the interval branch-and-bound-type methods. In particular, we discuss computing eigenvalues of interval matrices, use of algorithmic (automatic) differentiation, memory management techniques as well as advanced parallelization in both shared-memory and distributedmemory environments. Дан краткий анализ результатов исследований по применению интервальных методов для вычисления равновесия по Нэшу в непрерывных играх. Рассмотрены некоторые современные подходы к ускорению интервальных методов типа ветвей и границ. В частности, обсуждаются такие вопросы, как вычисление собственных значений интервальных матриц, использование алгоритмического(автоматического) дифференцирования, методы управления памятью, инструменты распараллеливания в средах с разделяемой и распределенной памятью.

EQUALS – a fast parallel implementation of a lazy language

1997 ◽  
Vol 7 (2) ◽  
pp. 183-217 ◽  
Author(s):  
OWEN KASER ◽  
C. R. RAMAKRISHNAN ◽  
I. V. RAMAKRISHNAN ◽  
R. C. SEKAR
Keyword(s):  
Shared Memory ◽  
Memory Management ◽  
Garbage Collection ◽  
Parallel Implementation ◽  
Parallel Machine ◽  
Functional Language ◽  
Parallel Performance ◽  
Strictness Analysis ◽  
Reference Counting ◽  
Sequential Implementation

This paper describes EQUALS, a fast parallel implementation of a lazy functional language on a commercially available shared-memory parallel machine, the Sequent Symmetry. In contrast to previous implementations, we propagate normal form demand at compile time as well as run time, and detect parallelism automatically using strictness analysis. The EQUALS implementation indicates the effectiveness of NF-demand propagation in identifying significant parallelism and in achieving good sequential as well as parallel performance. Another important difference between EQUALS and previous implementations is the use of reference counting for memory management, instead of mark-and-sweep or copying garbage collection. Implementation results show that reference counting leads to very good scalability and low memory requirements, and offers sequential performance comparable to generational garbage collectors. We compare the performance of EQUALS with that of other parallel implementations (the 〈v, G〉-machine and GAML) as well as with the performance of SML/NJ, a sequential implementation of a strict language.

Sign in / Sign up

Export Citation Format

Share Document