Now Available: Parallel Computing Server on Linux
- Apr 13, 2021
- Blog Post
- Developer Tools
STK Parallel Computing Server provides a set of client APIs and an infrastructure that enables applications to scale their execution across computing resources. Using the client APIs (available in .NET, Java, and Python), applications create jobs composed of tasks. Each task is a unit of work that gets executed in its own host process. Client applications submit jobs to a central coordinator. The coordinator schedules the jobs’ tasks for concurrent execution on one or more agent computers, which spawn the host processes. Through version 2.0, STK Parallel Computing Server only supported Windows platforms. However, many high-performance computing servers run Linux based operating systems.
In STK Parallel Computing Server 2.1, the coordinator and agent services were migrated to .NET Core, which made them capable of running on Linux systems. At the time, client applications that were to run on Linux were limited to using STK Parallel Computing Server’s Java and Python software development kits (SDKs).
Now in STK Parallel Computing Server 2.2, .NET Core client applications can also submit jobs to a Linux cluster. With cross-platform support across all three STK Parallel Computing Server SDKs — .NET, Java, and Python — you can now parallelize your custom STK Engine, ODTK, and STK Components applications on high-performance computing resources that run Linux.
For system requirements and installation instructions, visit the STK Parallel Computing Server Help.
To view the latest new features, check out the What's New in STK page.