HP Extended Cluster for RAC Continuous availability with the flexibility of virtualization

The complex tasks of managing and linking business priorities to the appropriate resource allocations
are handled by the highly integrated components of the VSE portfolio of solutions:
• HP Serviceguard and SGeRAC
• HP Partitioning Continuum
• HP-UX Workload Manager (WLM) and HP Global Workload Manager (gWLM)
• HP Instant Capacity and Temporary Instant Capacity
HP SGeRAC brings data protection, application availability, and ease of management to servers and
server partitions to create an enterprise cluster that delivers highly available application services to
LAN-attached clients. HP Serviceguard monitors the health of each node and responds to failures in a
way that minimizes or eliminates application downtime. When used in conjunction with SGeRAC, it
allows for rapid and transparent recovery from LAN and application failures, while still maintaining
scalability, data integrity, and configuration flexibility.
HP Partitioning Continuum components are the industry’s broadest range of hard and virtual
partitions, and they offer partitioning tools that provide resource flexibility, improved system
utilization, and lower costs in consolidated environments while maintaining the appropriate level of
isolation. The flexibility of HP Partitioning Continuum is automated with WLM and gWLM, which can
reallocate resources between partitions based on business policies and real-time requirements.
WLM provides automatic resource allocation and application performance management through the
use of prioritized service level objectives on a single system. It is the key to enabling applications to
be stacked. Goal-based resource management provides highly predictable response times for all
mission-critical applications, regardless of their location. The tight integration of HP Serviceguard and
WLM allows CPU and memory resources to be assigned to specific HP Serviceguard packages after a
failover. gWLM builds on WLM and provides similar resource management functionality. gWLM has
been designed for a centralized IT environment and employs a central management station to
facilitate resource management implementation across multiple systems.
Both WLM and gWLM guarantee computing resources to mission-critical applications but allow lower-
priority applications to utilize those resources when they are otherwise unneeded by the mission-
critical applications. In a SGeRAC implementation, one or more SGeRAC instances can be stacked
on the same infrastructure with the appropriate level of isolation while low-priority applications can
use the rest of the available computing cycles. For example, if a data center fails and the load
significantly increases on the remaining data center, it automatically transfers resources from the
lower-priority workloads to the Extended Cluster for RAC instances.
HP Instant Capacity is an innovative mechanism to rapidly respond to unpredictable business volatility
through the deployment of preconfigured systems that provide additional computing capacity. Clients
are charged when the additional resources are utilized. Instant Capacity provides a permanent
capacity increase while Temporary Instant Capacity allows additional capacity to be activated and
deactivated as required. Temporary Instant Capacity is integrated with WLM and gWLM 2.0+ to
automatically activate and deactivate capacity as needed. When used with SGeRAC, if one data
center goes offline and another data center picks up the load, additional CPU resources can
automatically be brought online to help maintain service levels.
HP Extended Cluster for RAC
The solution components bring together the best aspects of HP VSE, the HP suite of clustering
products, and Oracle RAC: comprehensive resource utilization, high availability, data integrity,
scalability, and reduced administration costs. Compelling on their own, these characteristics are
further complemented by the ability to share databases across multiple data centers, providing full
disaster tolerance and unprecedented levels of risk mitigation and cost effectiveness.
7