IBM Platform Computing

Maximize the Agility of your Distributed Computing Environment

 

IBM Platform Computing

Maximize the Agility of your Distributed Computing Environment


Platform LSF Wins Editors Choice Award

HPC Wire awarded IBM Platform LSF it's Editors Choice award at Supercomputing 2014

 

Platform LSF Wins Editors Choice Award


HPC Wire awarded IBM Platform LSF it's Editors Choice award at Supercomputing 2014

IBM Platform Computing

Maximize the Agility of your Distributed Computing Environment

IBM Platform Process Manager

Complex scripts are often used to automate complex, multi-step job sequences. But these scripts can be risky to modify, and may depend on the expertise of a few key individuals. IBM Platform Process Manager simplifies the design and automation of complex computational and analytic processes, capturing and protecting repeatable best practices.




The Challenge

Overwhelming complexity

Today, many engineering and analytic workflows are controlled by brittle, complex scripts that have evolved organically over the years. Often, few people fully understand the logic within the scripts making them risky and tedious to modify. The evolving complexity of distributed HPC and analytic environments, rapid data growth, and reductions in IT budgets, can result in multiple data silos, uneven processing, design cycle peaks and delayed results. IBM® Platform™ Process Manager helps increase the reliability of automated sequences of diverse workloads and improves the quality of results making workflows easier to maintain and helping make your business more agile.




The Solution

But it doesn’t have to be so hard

Platform Process Manager enables you to more easily design and automate computational and analytic processes, while capturing and protecting repeatable best practices. Workflow steps and dependencies are documented using an intuitive graphical interface, so you can automate lengthy, repetitive tasks that are prone to human error. An intuitive user interface makes it quick and easy to design workflows. And, when combined with IBM Platform Application Center, the complexity of these processes can be hidden behind an easy-to-use, process-specific interface. Administrators can then utilize documented flows without needing to understand the details, dramatically boosting their productivity, reducing run times and increasing overall process reliability.


Figure 1: Platform Process Manager automating a complex distributed workflow across multiple, heterogeneous data centers

Figure 1: Platform Process Manager automating a complex distributed workflow across multiple, heterogeneous data centers




The Benefits

Powerful, yet easy to use

Platform Process Manager is a powerful, easy-to-use interface for designing complex engineering computational processes that helps you:




The Specifics

Run critical engineering workflows with confidence

Platform Process Manager gives you the tools you need to run and manage business-critical workflows on a shared, distributed compute environment. Integration with the resource-aware workflow management capabilities of Platform LSF, and the ability to launch Platform Symphony and Hadoop MapReduce jobs as individual steps in a broader workflow provide for maximum flexibility. Platform Process Manager boosts productivity, reduces run times, improves process reliability and removes tedious manual steps. Technical users will appreciate the XML-based file formats and the rich command set that allows time-dependent jobs or flows to be triggered and managed using a scriptable interface as an alternative to the graphical environment.




This illustrates how flows and jobs can be managed easily through the optional Platform Application Center enabling complex multi-step workflows to be treated as individual units of work.

This illustrates how flows and jobs can be managed easily through the optional Platform Application Center enabling complex multi-step workflows to be treated as individual units of work.



The screenshot shows how multiple users can run multiple diverse workloads for execution on a grid at the same time, and cluster administrators have full control over how workloads run and the amount of resources they are able to consume.

The screenshot shows how multiple users can run multiple diverse workloads for execution on a grid at the same time, and cluster administrators have full control over how workloads run and the amount of resources they are able to consume.




Contact IBM


Connect with Platform Computing

Success Stories

Learn More

Training, Services & Support