Search Results for "network performance"

JPPF makes it easy to parallelize computationally intensive tasks and execute them on a Grid.

asyncoro is a Python framework for developing concurrent, distributed, network programs with asynchronous completions and coroutines. Asynchronous completions implemented in asyncoro are sockets (non-blocking sockets), database cursors, sleep timers and locking primitives. Programs developed with asyncoro have same logic and structure as Python programs with threads, except for a few syntactic changes. asyncoro supports socket I/O notification mechanisms epoll, kqueue, /dev/poll (and poll and select, where necessary), and Windows I/O Completion Ports (IOCP) for high performance and scalability, and SSL for security

Postgres-XL is a PostgreSQL-based scale-out cluster that handles both OLTP write intensive workloads as well as OLAP/BI type of workloads thanks to MPP parallelism. Postgres-XL is fully ACID and supports cluster-wide Multi-version Concurrency Control (MVCC) for global consistency and performance. There are commercial companies such as 2ndQuadrant, who provide support and consulting around Postgres-XL. Please visit http://2ndquadrant.com/en/resources/postgres-xl/ for more details.

Ganglia is a scalable distributed monitoring system for high-performance computing systems such as clusters and Grids. It is based on a hierarchical design targeted at federations of clusters. Supports clusters up to 2000 nodes in size.

Ichnaea is a set of tools that aid in collecting and tracking parameters and timings for parallel applications. The Performance Modelling Timing Module, PMTM, is a library that wraps system timing calls to abstract these from code developers and aid portability. It also has functionality to store parameters and print those, along with the timing information to a comma separated variable file. The Performance Modelling Analysis Tool, PMAT, is coming soon. This is able to read in and...

FastDHT is high performance DHT ( distributed hash table ) which based key value pairs. Storage use Berkeley DB, and network IO use libevent. It can store mass key value pairs such as filename mapping, session data and user related data.

NOTE: UBMoD is no longer actively developed and we encourage you to download Open XDMoD as a replacement for UBMoD. Open XDMoD has been developed to provide detailed information on resource utilization and performance for academic and industrial HPC centers. More information is available at http://xdmod.sourceforge.net/ UBMoD is a data warehouse and web portal for mining statistical data from resource managers in high-performance computing environments. UBMoD presents resource...

Scarlet is an extension for Atlassian JIRA, the leading Enterprise issue tracking and management software, allowing easy clustering of JIRA application instances, ensuring both high availability and linear scalability.

The CodeTime platform covers every aspect of parallel software from authoring, through distribution, to run-time. Its goals are: high programmer productivity; write once, run high performance anywhere; and wide acceptance.

RangeSim is a collection of classes for simulating distributed protocols for range search on P2P (peer-to-peer) network environments. It is based upon the PeerSim simulator, which offers easy configuration, extendability and high performance.

High Performance Computing test

Naanou is a 'clean-room' implementation of the published Chord protocol from MIT. It is a completely distributed hashtable P2P network, with provable high-performance characteristics.

MyVIA is sub-project of THVIA project. MyVIA is a high performance user-level communication library. It is a user-level net driver for Myrinet high performance network. Please visit our homepage and send email to me: chyyuu@yahoo.com OR chyyuu@sohu.com

Pluster is a cluster management software developed with python. It is easy to install and maintain. Pluster Provides cluster performance monitoring and remote OS installing. It's future will be an agent-based management system.

RingFS is a network distributed file system which is fault tolerant and has no centralized meta-data store. It spreads data over network storage servers and provides single namespace. The design of RingFS aims to build a huge data storage with redundancy, linear performance increase with rock-solid stability.