Wide Area Grid
Unified and Flexible Grid Management
Moab® Wide Area Grid is a powerful grid-workload management solution that provides unified scheduling, intelligent policy management, integrated resource management, data staging, distributed computing, and consolidated monitoring and management across multiple clusters. Wide Area Grid connects disparate clusters into a logical whole in a matter of minutes, extending the value of Moab®. It enables grid administrators through grid policies to have sovereignty over all High Performance Computing (HPC) systems while preserving control at the individual cluster.
Moab® Wide Area Grid allows organizations to consolidate reporting, synchronize management across policies, processes and resources, and optimize workload-sharing by staging and managing data across multiple HPC clusters. It delivers these services in a near-transparent way so users are unaware they are even using grid resources—they know only that they are more productive than ever before.
Solutions to Multi-Cluster Utilization and Management Challenges
Moab® Wide Area Grid helps manage and optimize multiple clusters by moving to a more efficient and productive grid environment and addresses the following key challenges:
- Multi-cluster sprawl inefficiency increases administrative burden and overhead costs as workload and policy management, monitoring, reporting, and planning must be done separately
- Unbalanced resource utilization slows job throughput and wastes resources as workloads wait to run on overloaded clusters and underutilized resources sit idle on other clusters
- Incongruent workload requirements, policies and SLAs across independent group- and organization-based clusters prohibit the sharing of workloads and makes unified workload decisions complex and hard to enforce consistently across multiple clusters
- Multiple resource managers make it impossible to rip and replace for grid management due to process and script investments, thereby inhibiting the effective scheduling and allocation of resources across clusters
- Multiple submission tools and credentials, found across different clusters makes it difficult to keep users productive
- Process data that resides within another cluster to quickly copy data from the external site to the existing cluster without reserving compute nodes or other resources
Maximize Multi-Cluster ROI
Moab® Wide Area Grid accelerates workload productivity and reduces management complexity by creating an optimized high performance computing grid environment with the following key benefits:
- Quick implementation with unified management across heterogeneous clusters that enables you to move quickly from cluster to optimized grid
- Faster job throughput and response times with policy-driven decisions and automation that accelerates complex workload scheduling to enable faster job start times and high throughput computing
- Cross-cluster optimization with flexible multi-dimensional intelligence engine that optimizes throughput, workload scheduling, and resource utilization at both grid and cluster level
- Ease of management with grid-wide interface and reporting tools that reduce management burden and costs through supplying a unified view of grid resources, status and usage charts, and trends over time for capacity planning, diagnostics, and accounting
- Scalable architecture to reliably support heterogeneous grids, high throughput computing, exascale and beyond
- SLA auto-enforcement on grid and local cluster levels, including, priorities, and resource sharing agreements across users, groups, and projects sharing grid resources
- Advanced administrative control allows various business units to access and view grid resources regardless of physical or organizational boundaries, or alternatively restrict access to resources by specific departments or entities
- Simplified use for both users and administrators through integrated job submission, grid-aware job arrays, job templates, optional user portal, and GUI administrator management and monitoring tool
- Improved cluster utilization with advanced workflow data staging with multiple transfer methods and types
Unified and Flexible Grid Management
Data Staging
Moab® Wide Area Grid’s capabilities also include grid data staging, which allow it to automatically manage data movement in context of when and where a job runs without blocking the resource. This is important as it simplifies a user’s experience and it avoids wasting resources when transferring large files. Moab® supports staging data to or from a shared file system on an unspecified cluster – resolved at job migration – in a grid configuration. Automating this process improves user productivity and optimizes resource utilization.
Moab® Wide Area Grid
Moab® Wide Area Grid provides automated decision control and management flexibility that real-world grid environments require. Its multi-dimensional intelligence engine accelerates workload output while balancing the complexities of grid, local cluster, and organizational workload priorities or policies. It can be flexibly configured to unify workload management policies into centralized management, provide centralized and local management policies, or integrate between local peer-to-peer cluster management policies. This ensures that service level guarantees and overall organizational objectives are achieved while utilization and results output across the unified grid environment are dramatically increased.
Flexible Grid Workload Management Architecture
Moab HPC Suite – Grid Option is architected to manage on top of the existing multiple job resource managers and other types of resource managers across the multiple clusters in your grid environment. This enables it to quickly provide unified policy-based workload scheduling optimization, and management. It extends the value of Moab HPC Suite Basic and Enterprise Edition by automating and unifying complex workload decisions and actions across multiple clusters. It utilizes the Moab intelligence engine and all of the data it integrates from the various resource managers to orchestrate the job and management actions through your existing resource managers based on policies. This makes it the ideal choice to integrate with existing and new systems and clusters as well as to manage your grid as it grows and expands in the future.
Moab HPC Suite – Grid Option can be architected in three flexible grid management configurations: centralized; centralized and local; or peer-to-peer grid policies, decisions, and rules. Its unique ability to manage multiple resource managers and multiple Moab instances makes this flexibility possible.
Resources
