Papers by Vladimir Vlassov
Lecture Notes in Computer Science, 2016
14-th IEEE International Conference on Peer-to-Peer Computing, 2014
2016 16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid), 2016
Page 1. Department of Microelectronics and Information Technology Royal Institute of Technology S... more Page 1. Department of Microelectronics and Information Technology Royal Institute of Technology Stockholm 2002 Performance Implication of Fine-Grained Synchronization in Multiprocessors Oscar Sierra Merino, Vladimir Vlassov, with Csaba Andras Moritz Page 2. Page 3. ...
Performance results of machines with fine-grain synchronization on individual lock-free data item... more Performance results of machines with fine-grain synchronization on individual lock-free data items (e.g., words), such as the MIT Alewife multiprocessor, illustrate the benefits of supporting fine-grain synchronization. The performance benefits are primarily the result of allowing a dataflow style of computation in programming models, and maximizing the exposed parallelism by minimizing the possibility of false dependencies caused by coarse grained synchronization.
This document presents the Distributed Programming Toolkit rNUT users' guide and reference pages.... more This document presents the Distributed Programming Toolkit rNUT users' guide and reference pages. The rNUT toolkit is an extension of the NUT System version 2.7 developed at KTH in June 1994. It consists of a library librnut and a daemon nutd. The library librnut contains low level NUT routines which provide support for building a system of collaborative NUT processes running on the Parallel Virtual Machine, PVM. To support interprocess communication, librnut contains routines for exchanging classes, scripts and objects between NUT processes. The daemon nutd is an executable code which is used to manage NUT process spawning and display connections. The synchronization and communication mechanism of rNUT is based on the Extended Dataflow Actor model, EDA, developed at KTH.
We use a functionally simple distributed file storage service to demonstrate a framework for deve... more We use a functionally simple distributed file storage service to demonstrate a framework for developing and managing self-* component-based applications for highly volatile Grid environments. The service replicates data for reliabil-ity, and provides for self-configuration, self-healing and self-tuning. The frame-work allows to develop application self-* behaviours as a distributed event-driven management application that is independent from application's functional code yet can interact with it when necessary. The framework provides a small set of abstractions that facilitate fault-tolerant application management, and fault-tolerance of the management itself. The framework uses and extends the self-* properties of the structured overlay network which it is built upon.
14 Th Ieee International Conference on Peer to Peer Computing, Sep 1, 2014
2015 IEEE 4th International Conference on Cloud Networking (CloudNet), 2015
Computers and Artificial Intelligence, 1998
An extensible object-oriented platform NUTS for distributed computing is described which is based... more An extensible object-oriented platform NUTS for distributed computing is described which is based on an object-oriented programming environment NUT, is built on top of the Parallel Virtual Machine (PVM), and hides all low-level features of the latter. The language of NUTS is a concurrent object-oriented programming language with coarsegrained parallelism and distributed shared memory communication model implemented on a distributed memory architecture. It differs from other languages of concurrent programming in the following: concurrent processes are represented by packages which are semantically richer entities than objects, inter-process communication is performed in terms of classes, objects, scripts and packages, using the EDA communication model; processes can be arranged into structured collections: grids which enable one to program data-parallel computations on a high level; sequential segments of programs can be synthesized automatically from specifications represented as classes using the program synthesis features of NUT. Examples of usage of generic parallel computing control structures PARDIF and PARARR are given.
ABSTRACT This article presents mEDA-2, an extension to PVM which provides Virtual Shared Memory, ... more ABSTRACT This article presents mEDA-2, an extension to PVM which provides Virtual Shared Memory, VSM, for intertask communication and synchronization. mEDA-2 consists of functions to access VSM and a daemon to manage parallel program termination. Access to VSM is based on the semantics of the EDA parallel programming model. The aim of developing mEDA-2 was to facilitate construction of parallel programs in PVM by providing a unified approach to message passing and shared memory models. 1 Introduction Most of the parallel programming environments, such as: PVM, P4, PARMACS, EXPRESS, MPI, etc., support the distributed memory programming model and are based on the message passing communication technique [1]. An alternative is the shared memory model which is characterised by its flexibility and ease of programming. In this article we present mEDA-2, an extension library for the Parallel Virtual Machine, PVM [2], which provides Virtual Shared Memory, VSM. This is a dynamic set of shared variables ...
2015 3rd International Conference on Future Internet of Things and Cloud, 2015
High-Performance Computing and Networking, 1999
A new coordination model is constructed for distributed shared memory parallel programs. It explo... more A new coordination model is constructed for distributed shared memory parallel programs. It exploits typing of shared resources and formal specification of a priori known synchonization constraints.
Computing and Informatics / Computers and Artificial Intelligence, 1998
An extensible object-oriented platform NUTS for distributed computing is described which is based... more An extensible object-oriented platform NUTS for distributed computing is described which is based on an object-oriented programming environment NUT, is built on top of the Parallel Virtual Machine (PVM), and hides all low-level features of the latter. The language of NUTS is a concurrent object-oriented programming language with coarse- grained parallelism and distributed shared memory communication model imple- mented on
2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing, 2015
Uploads
Papers by Vladimir Vlassov