HPC Systems Administrator

Job Title: HPC Systems Administrator (Windows + SQL Server)


Location: Buffalo, NY - Onsite


Mode: Fulltime

Job Description:


The resource is expected to own and independently execute all assigned responsibilities end-to-end, with minimal supervision.


Operate and support a Windows-based analytics platform running on Microsoft SQL Server and Microsoft HPC


Administer Microsoft HPC environments, including head nodes, compute nodes, and scheduling services


Monitor and manage compute-intensive workloads:


Job execution, queue backlogs, stuck or failed jobs


Expected vs abnormal CPU utilization on compute nodes


Maintain HPC cluster health:


Node availability, state management, and connectivity


Controlled maintenance, patching, and rolling reboots


Perform routine Windows Server operations:


OS patching, service monitoring, disk/CPU/memory/network health checks


Configuration of performance-sensitive OS settings


Administer Microsoft SQL Server instances:


Backup and restore validation.


Job monitoring, index/statistics maintenance


Transaction log and capacity monitoring


Troubleshoot cross-stack issues spanning:


HPC scheduler and compute nodes


Windows OS and services


SQL Server performance and availability


Manage service accounts, permissions, and security configurations across platform components


Support incident, change, and problem management activities


Maintain operational runbooks, health check procedures, and recovery documentation


Coordinate with application, infrastructure, and database teams during critical processing windows


Plan and execute migrations of Windows and SQL Server workloads between datacenters, including inventory. dependency analysis, cutover coordination, and post-migration validation


Assess SQL Server workloads for migration to Azure SQL (Database, Managed Instance. or Azure VM) and support remediation of compatibility gaps


Support data migration, cutover, rollback planning, and post-migration performance stabilization for both on-prem and cloud targets


Coordinate decommissioning of legacy servers and update runbooks, monitoring, and support procedures following migration