– CVS-RSA-BASE is the entry level service work package which is used to establish a base remote monitoring and administration framework. Through a secure remote access, this package is used to perform standard cluster status checks such as machine operation, interconnect subsystems, storage and application queuing. Monitoring is 24/7. All relevant notes concerning fixes, changes, updates and upgrades are documented, managed and shared via a change logbook.
– Secure Shell (SSH) is the standard protocol for remote cluster access. Secure Shell access is easy to set up, and can be quickly disabled when no longer required. The CVS-RSA-CONNECT service option can be added to accommodate other non-default remote access types. Supported environments include most IPSec, OpenVPN compatible, Cisco Anyconnect and TeamViewer implementations.
– The administration of end-user login accounts is managed via the CVS-RSA-ACCOUNTS service. Each package allows the creation, change and removal of up to 50 authenticated user accounts, and includes support for all main Linux NIS, LDAP and Kerberos ticketed protocols. Further CVS-RSA-ACCOUNTS packages can added to scale support for larger user communities. CVS-RSA-ACCOUNTS is also a starting requirement from which to scale to a full end-user help service using the CVS-RSA-HELPDESK service package.
– For cluster owners needing a comprehensive level of end-user support we highly recommend the addition of the CVS-RSA-HELPDESK service package. This package provides authenticated users with comprehensive first and second level help-desk style support. It also provides access to a wealth of information and forum contact via the ClusterVision online Service Portal. User accounts are required to be managed via the CVS-RSA-ACCOUNTS service package, and once in place up to 50 authenticated users can be supported within each CVS-RSA-HELPDESK package. Further CVS-RSA-ACCOUNTS and CVS-RSA-HELPDESK packages can be added to scale help-desk support to larger user communities. CVS-RSA-ACCOUNTS and CVS-RSA-HELPDESK are individually tailored services. Please contact our team for a discussion of your needs.
– This service package provides a remotely administered cluster environmental management. Building on the framework of the CVS-RSA-BASE service, this option enables ClusterVision engineers to monitor a range of environmental metrics and to diagnose and address any events triggered by exceeding the defined thresholds. Typical inclusions are integrated on-board sensors, IPMI sub-systems, APC environmental management kits, one-wire configurations and water cooling systems. This service includes both graceful and emergency shutdown processes, and can be optimised for specific hardware and energy efficiency requirements.
– As an up-scale service to the basic change-log provision of the CVS-RSA-BASE package, this service provides full monitoring and reporting of a cluster infrastructure. It includes full monthly aggregated reporting of all events and auto-matching of events to email alerts. Although not a mandatory requirement, it is recommended that the CVS-RSA-REPORT provision is enhanced with the additional security, auditing and management level reporting functionality of the CVS-RSA-SECURITY service package.
– This service package provides management, support and maintenance of the network systems in the cluster infrastructure. Typical actions include the creation and management of (r)STP, MLT, VLAN, bonding, trunking and stacking components. Each CVS-RSA-NETWORK package provides service for one supported network system, and can be easily scaled for complex or multi-network systems. Support for all common commodity network systems is available, including Avaya, Dell, Nortel, Mellanox, Flextronics, and Qlogic.
– This service package provides redundancy management and security updates for fail-over head-nodes, fail-over storage or fail-over front-end machines. Each CVS-RSA-FAILOVER package provides service for a single fail-over subsystem. Fail-over cover can also be scaled for parallel filesystems by the addition of the CVS-RSA-PFILESYSTEM service package.
– This service package provides maintenance, tuning, support, security updates, and rebuild/recovery for all of the common infrastructure filesystems. Typically supported systems include GPFS, LustreFS, FhGFS, PanFS, and ZFS (limited). Each CVS-RSA-PFILESYSTEM package provides remote service for one filesystem, and can be scaled as required by the addition of further packages. CVS-RSA-PFILESYSTEM can also be used together with CVS-RSA-FAILOVER to provide fail-over support for parallel filesystems. Details vary so please discuss your parallel filesystem requirements with our Services team.
– This service package provides remote administration for third-party tools, scheduling/queuing systems, and end-user applications. Provision varies according to individual needs, but typical actions include tailor-made installations, integration of installed applications into the HPC environment, and updates and bug-fixes for up to 10 installed user applications.
– This service package provides support and system administration for one major queuing and scheduling system combination. Supported systems include GridEngine, Torque, PBS Pro, SLURM, MOAB, Maui, LSF and OpenLava, and the service includes configuration, parallel environment management, and end-user sample scripts and instructions. Otherwise excluded services such as major version upgrades or changes to different queuing systems can often be incorporated by extending the service provided via the use of ClusterVision Prepaid Service Credits (CVS-CREDITS).
– The CVS-RSA-SECURITY service package provides peace of mind through a regular professional security auditing of the actively running cluster system. Actions include application testing for known vulnerabilities, intrusion and other malware scanning, and agreed processes for the safe lockdown, reporting and, where possible, cleaning of compromised systems.
– For additional system security, specifically for high integrity/availability data, we recommend the addition of the CVS-RSA-BACKUP service package. In addition to a permanently available factory default snapshot, this service package provides a periodic data health-check, and the creation and access to incremental backups for operating systems, management software and user applications and their associated data.