When comparing AWS PCS with TotalCAE, TotalCAE stands out as the superior solution for cloud HPC management. TotalCAE not only offers more convenient features but is also 2-10X cheaper.
What is AWS Parallel Computing Service (AWS PCS)?
AWS Parallel Computing Service (AWS PCS) is a managed service offered by Amazon Web Services (AWS) that assists customers in setting up and managing high-performance computing (HPC) clusters. The service allows users to run complex simulations and computations at virtually any scale. With AWS PCS, system administrators can create customized and managed Slurm clusters that align with their specific computing and storage requirements.
What is TotalCAE?
TotalCAE is a comprehensive cloud HPC solution designed to provide hassle-free access to cloud-based computing power. Through its intuitive, web-based portal, TotalCAE enables users to leverage the full potential of HPC without the need for extensive technical expertise. Its user-friendly interface allows users to run complex simulations and solve models with just a few clicks.
Our cloud HPC solution is managed entirely within your cloud subscription and is compatible with all major cloud providers such as AWS, Azure, Google Cloud Platform (GCP), and Oracle Cloud Infrastructure (OCI), giving you flexibility in your cloud vendor choice.
TotalCAE is a full service, the user can write their own scripts and install their applications but we can also install, debug, maintain, and update their applications and take care of the entire environment. By handling all aspects of HPC management, TotalCAE makes it easier for organizations to focus on their core simulation or computation work rather than the complete end to end infrastructure.
AWS PCS vs TotalCAE
Here are the main differences between AWS PCS and TotalCAE.
Management
Amazon’s PCS allows you to login to a cluster, and submit a script to run an application. You can turn on a desktop to view results. AWS PCS requires IT to support and manage the setup, AWS will update the scheduler periodically, and leaves it to users to deal with their applications, application licensing, and the rest of the integration needed for a production solution.
You could also do this with TotalCAE on AWS, but our clients do not want to do all the remaining low-level work that is required. Having a shell on a cluster is the most minimal setup and not “production ready” for most workloads. Our solution is fully managed, we take care of everything from start to finish, and not just update the scheduler from time to time as PCS does.
Service Options
In terms of service options, Amazon’s Parallel Computing Service provides cloud-based HPC shell solutions for a user to implement and manage the rest of the solution. But it is limited to AWS’s ecosystem and its requirements. TotalCAE, however, is completely turnkey end to end, including applications, and offers more flexibility by integrating with multiple cloud providers, including AWS, Microsoft Azure, GCP, and more.
Users can also opt for on-premises solutions with TotalCAE, ideal for those with consistently high workloads, or they can adopt a hybrid approach, pairing on-premises infrastructure with their preferred cloud provider for enhanced cost-efficiency, customization, and control.
Features
Amazon’s solution has a range of features that make it easy to run and scale your HPC, however, TotalCAE includes the same features and much more. Here’s a quick comparison:
AWS Parallel Computing Service’s features include:
- API’s to create services on top of PCS
- Dynamic resource provisioning and scaling for non-licensed applications
- Operating System and Scheduler Software updates
- Supports custom AMIs the client builds
- Supports AWS HPC services and instance types
- Runs in your AWS environment
TotalCAE’s include all of the above, in addition to (but are not limited to):
- Hundreds of application templates built-in with application-aware job actions such as check pointing, examining job results, and other turnkey setups to make it easy to add in applications with high level integration as needed
- License aware scheduling to schedule around your application licenses
- Allows you to submit from Windows or Linux to avoid writing scripts through our portal, command line portal, or via API
- Integrated billing system
- Capacity planning tools
- Auto-stop cloud desktop
- Web portal for job submission
- Command line portal for scripting
- Single view of all of your AWS HPC clusters with TotalCAE MultiHPC
- API
- The basic “scheduler script” integration AWS supports
- Runs in your AWS environment, on-premises, and in any other clouds
- Tools to orchestrate data to and from on-prem to the cloud and back. Supports hybrid scenarios, along with multiple cloud enablement, all from a single interface and vendor
- Application-aware level monitoring to alert users to jobs that are having problems
- Capacity planning, job monitoring, and analytics
- Tools to manage job spending by the job and trace the simulations
- Simple Simulation Data Management
TotalCAE covers all the features required for engineers to get work done in just a few clicks, not just a basic cluster setup with users left to fill in the gaps. Learn more about TotalCAE’s features.
Pricing
When it comes to pricing, both PCS and TotalCAE on AWS operate within your AWS account, meaning that, with either PCS or TotalCAE, you pay the direct costs of your cloud infrastructure to AWS.
However, TotalCAE offers a fully managed service that is significantly more cost-effective. Our service is anywhere from 2-10X cheaper than PCS due to the additional charges AWS charges for its PCS service.
TotalCAE Use Case: Lucid Motors + Managed HPC Cloud
Faced with needing to design and deliver multiple award winning next generation EVs, Lucid needed timely crash simulations for vehicle design with both on-premises and cloud resources. Lucid Motors adopted TotalCAE’s cloud-enabled simulation platform on AWS. This solution, featuring TotalCAE’s Hybrid Multi-HPC enablement and AWS infrastructure, seamlessly integrated with their workflows and significantly accelerated simulation processes, enabling faster design iterations and improving overall efficiency.
“TotalCAE is simple to use, straight forward, and enables us to focus on engineering and not HPC complexity.” – Guang Dong, CAE Crash Lead — Lucid Motors
Harness the Power of HPC With TotalCAE
TotalCAE’s fully managed service is end to end from job start to job returns with 1-hour help desk, and one vendor. We don’t leave all the remaining things for the user to implement.
Contact us today to get started or to learn more about our managed HPC solutions.