World’s 2nd largest heavy equipment manufacturer accelerated digital innovation
Fovus empowered us to easily migrate 100% of our HPC workloads to the cloud. Fovus intelligently optimizes our HPC strategies and maximizes our license utilization, improving our digital innovation productivity by +30%.
Malek Abdel-Fattah
IT Director, Engineering and Manufacturing Applications
Single
command deployment
Intelligent
license usage management & balancing
Optimal
HPC strategy
Continuous
time-cost improvement
100%
HPC workloads migrated to cloud
30%
more innovation productivity
=
$6M
employment cost saving/100 engineers/year
+
Faster
time-to-market
Single
command deployment
Intelligent
license usage management & balancing
Optimal
HPC strategy
Continuous
time-cost improvement
30%
more innovation productivity
=
$6M
employment cost saving/100 engineers/year
+
Faster
time-to-market
About Komatsu
Komatsu Ltd. is a Japanese multinational company, the world’s second-largest construction, mining, and industrial equipment manufacturer. Founded in 1921, it is known for its advanced technology, high-quality machinery, and global presence.
Industry
Heavy equipment (construction, mining, and forestry), defense technology.
HPC Workloads
Structural analysis, multi-body dynamics, discrete element method, system simulation, etc.
Pains
- Engineering teams waste valuable time managing cloud logistics and exploring HPC strategies (hardware choices, cluster configurations, CPU/GPU parallelism, etc.), leading to about 30% innovation downtime.
- A suitable HPC strategy for workload A could be bad for workload B. Without proper HPC strategies for various workloads, the teams often experience long simulation runtime and high cloud costs.
- Many users or teams share the same software licenses, making it challenging to coordinate license usage across a large number of users.
- License competition and over-usage by individuals cause work downtime or failed simulations without user awareness. This leads to low work efficiency and poor license utilization.
- Software license availability is highly dynamic, making it cumbersome to provision the right amount of cloud infrastructure or scale it accordingly.
Solution
Serverless HPC experience, ease-of-everything
- Fovus uses AI to automatically orchestrate all the cloud logistics and optimize the HPC strategies so the teams can finally focus on innovation.
- The teams need only deploy their HPC workloads in a single command via CLI or with just a few clicks using the web UI, and Fovus takes care of everything else.
- Easy to migrate and integrate with existing workflows.
- Easy to deploy, scale, optimize, and sustain.
Intelligent job queueing & license usage balancing
- The administrator created custom license queues, each with different license usage constraints that define the desired license usage patterns against day and time.
- Jobs submitted are automatically assigned to the proper queue.
- Each queue intelligently ranks, queues, and deploys HPC jobs according to the software license availability dynamics, current license usage statistics by users, workload license consumption profiles, and the admin-defined license usage constraints to auto-balance license usage among different users to maximize the teams’ work productivity.
- No more hassles in coordinating or competing for licenses.
License utilization insights
- Fovus gives the administrator business insights on how well different users and software utilize the licenses over time.
License-constrained auto-scaling
- Fovus auto-scales the cluster size and the tasks running in parallel in each deployed job according to the software license availability dynamics and the admin-defined license usage constraints to maximize the overall utilization of software licenses across the teams.
- This avoids unnecessary cloud usage and eliminates any license-induced simulation errors.
Free automated benchmarking
- Fovus auto-benchmarked each type of Komatsu team’s production CAE workload for free.
- Such strategy space explorations reveal how computing parallelism and different hardware configurations impact the performance and cost of their CAE simulations.
- Such benchmarking data provides the foundation for optimizing the HPC strategy tailored to each Komatsu’s CAE workload.
AI-optimized HPC strategy
- Based on the custom benchmarking data, Fovus auto-determines the optimal HPC strategy to minimize the total runtime and cloud cost for each of Komatsu’s CAE jobs.
- This ensures the time-cost optimality of CAE simulations with zero user effort needed.
Continuously improve HPC strategy
- Fovus auto-updates the custom benchmarking data and continuously improves the HPC strategies for each Komatsu’s CAE workloads as cloud infrastructure constantly upgrades.
- This sustains the time-cost optimality of CAE simulations as cloud infrastructure quickly evolves.
Results
- 100% of production CAE workloads easily migrated from on-premises to the cloud.
- 30% higher team innovation productivity achieved.
- Delivered simulation results much faster and cheaper
- E.g., multi-body dynamics design of experiments (DOE) simulations runs 3x faster at a 4x lower cost than the prior solution.
- E.g., structural analysis simulation runtime reduced by 4.5x
- Nearly 100% of overall license utilization when jobs are present.
- Largely reduced waiting time per user.
- Enjoy the freedom from cloud management and license coordination hassles.
- Enjoy the time-cost optimality of CAE simulations and innovation efficiency.
- Ability to consistently leverage state-of-the-art hardware technology to turbocharger digital innovation and stay ahead of the curve.
Other case studies
Stealth Series-B Startup
Success Story
Trailblazing Biotech startup accelerated DMTA cycles at reduced cloud expense
Success Story
Leading discovery service provider accelerated DMTA cycles and project delivery at reduced costs
