| | |

Announcing Rescale Labs: Creating New Innovations for R&D Computing

Rescale introduces a new program to make internal tools available to engineers and scientists, helping R&D teams overcome computational challenges and unlock new innovations

If you use or support high performance computing (HPC) systems to conduct research or engineering, then you understand how specialized each system and workload can be. 

To help organizations address the complexity of HPC, Rescale has made it far easier to set up and run R&D computing tasks. By automating many of the steps for connecting to hardware, managing software, and tracking the workflows of data-intensive compute jobs, Rescale helps organizations dramatically accelerate their innovation efforts and drive the digital transformation of their research and engineering programs. 

As part of our ongoing platform development, we have documented patterns of common challenges facing our customers in their HPC operations. As a result, we’ve built many internal innovations that help our customers improve how they manage their R&D computing. 

In working alongside the brightest minds in science and industry, we’ve partnered with our customers to bring even greater scalability, performance, and reliability to HPC for research and engineering.

And now we’re excited to share these innovations publicly through Rescale Labs. The goal of Rescale Labs is to promote collaboration and new solutions that help our customers address their biggest R&D computing challenges.

Introducing Rescale Labs

Beginning today, Rescale will share information and resources with HPC practitioners and citizen application developers that will help them adopt new tools developed by our technical field teams. These are not stand-alone products. They are designed to offer quick enhancements that deliver greater speed, accuracy, and efficiency on the Rescale HPC platform.

These tools are complementary for use by Rescale customers and accessible on the Rescale Labs GitHub and in the Customer Knowledge Base

These capabilities and documentation should be considered preliminary as we focus on shipping more new innovations at greater speed. Depending on customer adoption and feedback, we’ll incorporate them into future product enhancements. 

We are looking forward to learning from our customers and finding new ways to make HPC more efficient, more secure, and more effective.

Access Rescale’s Internal and Experimental Innovations

We are building an open community to engage customers in ongoing testing, brainstorming, and, eventually, co-developing computational tools that help others in the community. To kick off the initiative, we are debuting two new innovations that can be implemented quickly and easily.

Job Analyzer: Monitor Cloud Hardware Utilization for Any Job

Job Analyzer provides visual and near real-time data on hardware utilization including CPU, GPU, storage, and memory for each compute node. 

These compute cluster metrics can identify resource constraints and diagnose suboptimal configurations to troubleshoot issues faster and reduce the risk of costly overruns. 

R&D and IT personnel can also use Job Analyzer to identify optimal hardware-software configurations that improve the performance and efficiency of software. 

After developing this capability internally, many of our customers requested this data to enhance their routine computing operations. Now it’s available in the product for anyone to use.

Job Analyzer monitoring a computational fluid dynamics workload with efficient resource utilization.

Rescale Web Apps: Democratize Access to Advanced R&D Workflows

Rescale Web Apps are a unique method for packaging complex R&D applications for simple, on-demand browser-based access. Powered by Rescale Publisher, Web Apps make it easy to deploy in-house or third-party applications using the latest compute architectures and 1,000+ pre-installed simulation and AI/ML software from Rescale’s software catalog. The simplicity of the application setup and access allows a broad set of users to benefit from advanced computational methods in science and engineering.

Web Apps offer intuitive user interfaces that provide automated simulation workflows developed by experienced simulation engineers. Simulation app developers now can focus on the user experience without worrying about configuring their applications to run reliably in various clouds or architectures. R&D leaders and citizen developers can empower non-expert team members to carry out simulation tasks consistently and efficiently, such as a field engineer testing design performance from a basic laptop or tablet. 

Web apps can be used for a variety of applications, such as:

  • Training and evaluating neural network architectures for image classification by non-AI specialists. 
  • Monitoring simulation progress by visualizing preliminary results for commercial and in-house CAE solvers. 
  • Simulating specific, recurring engineering and design scenarios, such as a solar panel array design app (shown below; available in the GitHub repository).

Deploying a custom solar panel array simulation app from the Rescale software catalog, then accessed interactively via in-browser graphical user interface.

Join the Cutting Edge of Computational Engineering

We are thrilled to kick off the Rescale Labs Initiative and look forward to future releases.

To learn more about Rescale Labs, you can visit our new dedicated GitHub repository to view projects as they become available. 

Existing Rescale customers can also learn more about how to use these innovations to resolve common issues they may encounter in our existing Customer Knowledge Base

We will continue to share new developments in the Rescale Blog (Tag: Rescale Labs). Stay tuned for more announcements to come!

Author

  • Garrett VanLee

    Garrett VanLee leads Product Marketing at Rescale where he works closely with customers on the cutting edge of innovation across industries. He enjoys sharing customer success stories, research breakthrouths, and best-practices from Rescale engineers, scientists, and IT professionals to help other organizations. Garrett is currently focused on the convergence of supercomputing, HPC, and AI simulation models and how these trends are driving discoveries in science and industry.

Similar Posts