Vega 3 Compute Pool Manager (CPM)
Manage and Monitor Compute Resources
The Azul Compute Pool Manager (CPM) is a flexible, policy-based management interface that enables system administrators with comprehensive command and control of compute pool resources. CPM, through an easy-to-use web-based console, dynamically allocates, redistributes compute resources to guarantee service levels, resource commitments, availability, and priorities based on business goals. CPM offers a consolidated view of all the resources in a pool and allocates resources in the compute pool based on default or administrator defined rules-based polices.
The Compute Pool Manager also monitors applications and appliances, allowing administrators to change resources allocated to running applications, log events and utilization information, and monitors the health of the compute appliance hardware. SNMP monitoring capabilities have been built-in so that CPM can easily interface with existing network management products.
Key Features
Extraordinary Appliance Ease of Use. Provides a powerful, easy-to-use web-based console enabling administrators to easily configure compute pools, define the resource policy, and monitor resource utilization and load distribution. Seamlessly integrates with existing data center management software tools.
Comprehensive Policy-based Resource Control. Provides command and control of compute pool resources; minimizes hands-on administration by providing default policies for adaptive resource allocation, intelligent workload distribution, interference protection and high availability.
Real-time Utilization Tracking and Billing. Track resource usage to support capacity planning, internal billing, and monitoring of service level agreement (SLA) compliance.
Enterprise Availability
Improves application availability, performance, and reliability to enterprise data centers running mission-critical applications in a heterogeneous server environment. Redundancy requirements and resource guarantees can be specified to protect against single points of failure and provide for high-availability, while unallocated resources can be efficiently utilized according to application priority.
Strict Authentication and Authorization Security
Integrates into existing authentication and authorization directories through RADIUS and LDAP focused on network-wide centralized authentication and accounting services.
Highly Available Resource Pooling Capabilities
Compute Pool Manager allows multiple compute appliances to be combined to form large capacity, highly available shared resource pools. Policy-based admission control and resource configuration simplifies application governance and routine adjustments.
Key Features:
Managed Resource Pooling. Pools consisting of multiple compute appliances can be configured for extended. shared capacity. Multi-level pools can be created to direct applications to any appliance in a pool, or to a preferred subset. Pools can be dynamically reconfigured for online expansion or re-assignment of compute capacity.
Fault-tolerant Pool Management. The compute pool remains available to applications and administrators upon any appliance going offline for any reason. Management elements fail over transparently, and launching against the pool as well as management of remaining appliances and applications is virtually uninterrupted.
Policy Based Resource Controls. User-defined policies dictate which applications are allowed to run, and reserve memory and minimal processor cores to guarantee availability for applications. Intelligent placement of launched instances on appliances maximizes system utilization and availability for additional workloads.
Policy Based Load-Balancing and Redundancy. Placement of applications on appliances can be directed to optimize load balancing across compute appliances within the pool and to guarantee application redundancy.
Workload Management Capabilities
The workload management capabilities elevate the control over the entire pool of compute appliances and its resources to mainframe-like levels, while keeping the same simple usage principles. Strict control and monitoring of resource allocation and use allow delivery of compute as a utility.
Key Features:
Policy Based Service Level Guarantees. Compute resources are allocated dynamically based on policies, guaranteeing consistent service levels, high availability, and dynamic alignment of resources with shifting business priorities.
Priorities. Allocation of processor cores to competing applications when use levels exceed guaranteed minimums can be strictly controlled according to the specified priorities of applications.
Reservations and Resource limits. Minimum resource levels can be guaranteed through reservations for critical applications, at all times. Maximums can be set for the number of concurrent application instances allowed per appliance and for levels of memory and processor cores allocated to individual applications.
Sparing. Memory and processing resources can be set aside on compute appliances to guarantee that capacity is available to restart critical applications in the event of compute appliance failures.
Usage Data Records. The CPM Data Manager enables real-time accumulation of usage data for applications, compute appliances, and compute pools, allowing tracking of capacity utilization trends and billing for application use of compute pool resources.
Remote Control and Integration. A Java based API enables customized integration with enterprise management and orchestration frameworks. The API enables centralized management of CPM resource policies, scripting for automatic compute pool management, disaster recovery handling, and retrieval of resource usage information. As a result, IT organizations can simplify system management and speed time-to-deployment for new installations.
