Make Appointment

SPOLS (Scalable Parallel Optimized Linux System)

An integrated solution for HPC cluster management.

High Performance Computing (HPC): An Overview

High Performance Computing (HPC) refers to a branch of computing focused on developing computational systems capable of executing complex numerical simulations and analyzing large volumes of data at extremely high speeds.

These systems, often called supercomputers, consist of thousands or even millions of processors working in parallel to solve problems that would take a prohibitive amount of time on a single computer.

High Performance Computing

SPOLS: The Integrated Solution for HPC Cluster Management

SPOLS is a comprehensive software suite designed to simplify the management and use of HPC clusters, ensuring maximum reliability and availability. The solution is structured on three distinct layers, each with specific functionalities:

Backend

  • Hardware Interface: The SPOLS backend directly interacts with the cluster's hardware resources, ensuring efficient and optimized management.
  • spolsadm: This application is responsible for the installation, configuration, and maintenance of the HPC cluster, including setting up HA (High Availability) controller nodes to ensure maximum system availability.
SPOLS Backend
SPOLS Management

Management

  • Software Interface: The management layer interfaces with various cluster management software such as Slurm, Munge, and MPI.
  • spolsmgm: This application provides a unified interface for user management, resource scheduling, and cluster activity monitoring. It allows for configuring access policies, defining resource quotas for users, and monitoring system status.

Frontend

  • User Interface: The frontend serves as the contact point between the user and the HPC cluster.
  • spolscmd: This command-line application allows users to submit jobs, monitor execution status, and manage job priorities and assigned resources.
SPOLS Frontend

Web Interface

In addition to the command-line interface, SPOLS offers an intuitive web interface that allows users to perform the same operations graphically. The web interface also integrates with OOD (Open OnDemand), a web portal that provides a richer user interface for managing resources and activities.

SPOLS Web Interface

Key Benefits

  • High Availability (HA): Minimizes downtime in case of hardware or software failures.
  • Scalability: Allows for adding new nodes to the cluster to increase computing capacity without interrupting service.
  • Load Balancing: Distributes the workload evenly among active nodes, improving cluster performance.
SPOLS High Availability Diagram

Monitoring

  • Comprehensive Monitoring: Offers a complete view of the cluster's status.
  • Flexibility: Zabbix offers a wide range of monitoring functionalities and can be customized to meet specific customer needs.
  • Centralization: Centralization of data simplifies management and analysis.
SPOLS Monitoring
#

REQUEST A FREE QUOTE

Get answers and advice from whoever you want.

Contact Us