About Advanced Resource Connector

The Advanced Resource Connector (ARC) middleware integrates computing resources (usually, computing clusters managed by a batch system or standalone workstations), making them available via a secure common Grid layer. Conceptually, ARC provides an edge service to batch systems and temporary file cahes. Through this service, called ARC Compute Element (ARC-CE), users can submit computational tasks to a variety of batch systems in a uniform manner, using a generic job description language. One of the key features of ARC is that it handles input and output data movement on the server side, thus optimising bandwith consumption and increassing overall efficiency of computing resource consumption. ARC is complete with a client toolkit, which relies on resource information in order to select a best matching facility. Information about facilities is collected by dedicate probes an published via ARC-CE. Only authorised users can submit tasks, and authorisation relies on X509 proxy certificates and the concept of Virtual Organisations.

Complete documentation, including component descriptions, installation details, usage instructions, articles and presentations, can be found at the documentation page.

Until recently, ARC relied on largely proprietary technologies based on GridFTP for most communications, and LDAP for information publishing. Today, Web Service based interfaces are being actively promoted. For backwards compatibility with deployed services, both pre-WS and WS-technologies are available in ARC releases.

Background

The ARC middleware by NorduGrid (in the past called NorduGrid middleware) is a software solution that uses Grid technologies to enable sharing and federation of computing resources distributed across different administrative and application domains. ARC is used to create Grid infrastructures of various scope and complexity, from campus to national Grids. It is distributed under the Apache v2.0 license.

In a nutshell, ARC is:

A general purpose, Open Source, lightweight, portable middleware solution
A reliable production-quality and scalable implementation of fundamental Grid services
Facilitator of cross-organisational distributed computing infrastructure solutions
Strongly commited to open standards and interoperability
In continuous use since 2002

Selected ARC components

ARC design principles

User- and performance-driven development:

start with something simple that works for users and add functionality gradually

Simplicity is a key in planning ARC development
No single point of failure or bottlenecks - stable by design
Non-intrusive and easy to maintain on the server side

Tools and utilities respect local policies, in particular, security policies

Flexible, powerful and simple on the client side
Provide a lightweight, stable, robust and fault-tolerant system

ARC development

ARC development is coordinated by NorduGrid, with contributions coming from many projects, most notably from NeIC, eSSENCE, as well from EU projects KnowARC, EMI, and others.
Current goal is to gradually upgrade established and well performing and yet technologically inflexible components by improved and extended ones, based on Web-Service architecture
The new ARC components features:

Service Oriented Architecture
Modular structure
Self-sufficient core components
Exploitation friendly Open Source license (Apache 2.0)
Portability: runs on all Linux variants, with most components also portable to Windows and Mac OS
Integration into Fedora (EPEL), Debian and Ubuntu public repositories

Internal structure of ARC components

All communication between client and server side can be done via Web services
User clients are developed using the client libraries
ARC offers both a command line interface (CLI), as well as a GUI
Python and Java bindings allows easy development of application-specific clients
On the the server side the HED (Hosting Environment Daemon) handles the communication between services and the outside world
HED dispatches messages to services which in turn communicate with backends and through them local batch systems

Pre-Web-service ARC

Pre-WS ARC provides a reliable implementation of the fundamental Grid services, such as information services, resource discovery and monitoring, job submission and management, brokering and data management and resource management, using customized and largely proprietary approaches. Most of these services are provided through the security layer of the GSI. The middleware builds upon standard Open Source solutions like the OpenLDAP, OpenSSL, SASL and Globus Toolkit^® (GT) libraries. Pre-WS ARC provides solutions essential for creation of a distributed computing infrastructure: the A-REX, the information model and providers (both NorduGrid an GLUE2 schemas), User Interface and broker (a "personal" broker integrated into the comman-line interface), extended Resource Specification Language (xRSL), and the monitoring system.

Pre-WS ARC main components are:

Grid services running on the resources: the A-REX and the information services. Grid jobs are submitted to the cluster through gridftpd and a separate session directory is created for each job. The Grid session directories are made available through the gridftpd server during and after job execution. The A-REX is a service running on a computing resource taking care of jobs, session directories and the input data cache area. Information services are implemented as probes populating the information database stored in OpenLDAP backends.
Indexing services for the resources and data: a special usage of an OpenLDAP database allows to build a hierarchical mesh of Grid-connected sites. For data indexing, ARC middleware relies on 3rd-party solutions, such as e.g. the Rucio service. ARC client tools and the A-REX are capable of interacting with these services.
Clients making intelligent use of the distributed information and data available on the Grid. ARC comes with a light-weight command line client, the User Interface. The ARC User Interface is a set of tools to submit, monitor and manage jobs on the Grid, move data around and query resource information. The User Interface comes with a built-in broker, which is able to select the best matching resource for a job. Grid job requirements are expressed via xRSL. Another special client is the Grid Monitor, which uses any Web browser as an agent to periodically query the distributed information system and present the results as a set of inter-linked Web pages.

ARC is designed to be a scalable, non-intrusive and portable solution. The development is user- and application-driven, with the main requirements being those of performance, stability, useability and portability. As a result of this approach, the standalone client is available for a dozen of platforms and can be installed in a few minutes. The server installation does not require a full site reconfiguration. The middleware can be built on any platform where the external software packages (like GT libraries) are available. While being deployed on a large production Grid and being used by real users, the middleware is naturally undergoing continous real life tests.

Last, but not least, ARC is designed with international users in mind: the client tools are available in several languages. The Grid Monitor is translated to more than 10 languages, and the User Interface is available not just in English, but also in Russian and Swedish. Volunteers to provide more translations are always welcomed!

User support, feature/bug reports, documentation

User support and site installation assistance is provided via EGI's GGUS request tracker. In addition, NorduGrid runs several mailing lists among which nordugrid-discuss is a general forum for all kind of issues related to the ARC middleware.

NorduGrid deploys the Bugzilla problem tracking system. Feature and enhancement requests, as well as discovered problems, should be reported there.

Research papers, overview talks, reference manuals, user guides, installation instructions, conference presenations, FAQ and even tutorial materials can be fetched from the documents section.