The TotalSim team was on the ground in Oshkosh for the Vertical Flight Society’s 20th Annual Electric Aircraft Symposium (EAS 2026), held July 18–19 at the University of Wisconsin–Oshkosh — the weekend before EAA AirVenture. EAS remains one of the best gatherings in advanced air mobility: intimate, technically deep, and packed with the people actually designing, building, and certifying the next generation of electric, hybrid-electric, and hydrogen-electric aircraft. Sessions covered the full spectrum — eVTOL, eCTOL, eSTOL, propulsion systems, infrastructure, and the latest regulatory developments including MOSAIC and the eVTOL Integration Pilot Program. We exhibited and had excellent conversations with developers, propulsion suppliers, and fellow simulation and engineering partners. It’s always valuable to be in the room where the real technical discussions happen, especially as the industry moves from concepts into more rigorous validation and certification work.
The TotalSim team was also on the ground in Oshkosh for EAA AirVenture, immersing ourselves in the full spectrum of aviation—celebrating its rich history, experiencing today’s cutting-edge aircraft and technology, and connecting with the innovators shaping its electric and advanced air mobility future. From classic warbirds to the latest eVTOL and experimental designs, the energy and passion on the flight line never fail to inspire. We’re proud to support the industry with high-fidelity simulation as aviation continues to evolve. Looking forward to building on the connections made and continuing to support the industry with high-fidelity CFD, simulation, and engineering solutions as aviation continues to evolve.
The TotalSim US Team
The Rise of Containerized Data Centers: Benefits, Challenges, and the Road Ahead
Shipping container data centers—prefabricated modular units that pack servers, storage, networking, cooling, and power systems into standard ISO steel containers—are transforming how organizations deploy IT infrastructure. Factory-built and pre-tested, these mobile solutions can go live in weeks or months rather than years, making them ideal for edge computing, disaster recovery, temporary capacity, and remote operations.
Key Benefits
The advantages are compelling. Rapid deployment stands out: pre-configured units arrive ready to connect to power and networking, slashing time-to-market dramatically—sometimes to as little as 90 days. Portability and flexibility allow easy transport to remote sites, disaster zones, mining operations, oil platforms, or temporary project locations.
Scalability is another major plus. Organizations can add capacity incrementally by deploying additional containers as demand grows, avoiding the over-provisioning common in traditional builds. Many designs emphasize energy efficiency with high-density configurations, hot-aisle containment, advanced cooling, and low Power Usage Effectiveness (PUE) ratings—often reaching 1.1–1.3 in optimized setups. This translates to lower operational costs and a reduced carbon footprint. Cost efficiency extends to construction savings, standardized components, and predictable procurement.
Durability is built-in: weatherproof, shock-resistant, and dustproof containers perform reliably in harsh environments. Leading providers include Schneider Electric, Dell Technologies, Hewlett Packard Enterprise (HPE), Microsoft (with modular Azure deployments), IBM, Delta, and others offering 20ft or 40ft solutions tailored for various power densities.
Real-World Use Cases
Common applications range from 5G edge networks and AI training at the edge to backup infrastructure and scientific research stations. Recent examples include AI startups deploying high-density GPU containers in desert conditions and telcos using them for low-latency edge computing.
Challenges to Consider
Despite the upsides, challenges exist. Space constraints make containers cramped for technicians, complicating maintenance and repairs. While individual units offer redundancy, larger setups may face single points of failure if not designed carefully. Physical security in remote or distributed locations can be harder to maintain than in centralized facilities, increasing vulnerability to tampering or environmental threats.
Integration complexity, specialized maintenance requirements, and potential limitations in concurrent maintainability are other hurdles. Upfront costs for high-end configurations can be significant, and managing a distributed fleet adds IT operational overhead. Cybersecurity, power reliability in off-grid scenarios, and compliance in varied regulatory environments also demand attention.
The Bottom Line
Containerized data centers deliver speed, agility, and efficiency in an AI-driven world hungry for compute. As the market grows rapidly, they’re not a universal replacement for traditional facilities but a powerful complement—especially for hybrid strategies balancing core data centers with edge and modular capacity. Organizations that plan thoughtfully around maintenance, security, and integration will reap the greatest rewards.
What’s New in Siemens Simcenter
STAR-CCM+ 2606
Simcenter STAR-CCM+ 2606 (released July 1, 2026) continues Siemens’ three-release-per-year cadence and delivers a strong focus on GPU acceleration, multiphysics depth (especially electromagnetics + thermal), meshing productivity, post-processing, and AI-assisted workflows. The release builds on earlier GPU foundations and expands native GPU support across a much wider range of physics, while introducing a new Finite Element Magnetic Field solver aimed at electric machine simulation.
GPU Acceleration: Broader Physics Coverage and Significant Speedups 2606 is primarily positioned as a GPU acceleration release. GPU-native solvers now cover (or expand coverage for):
Eulerian Multiphase (EMP)
Lagrangian Multiphase (LMP – Phase 1)
Fluid Film and Dispersed Multiphase (DMP)
Mesh-based Discrete Element Method (DEM)
Complex Chemistry
Eddy Break-Up (EBU)
Virtual Body / DFBI cases
Additional multiphase options (e.g., certain MMP multicomponent evaporation/condensation and VOF flash boiling)
Reported benchmarks versus CPU clusters range from roughly 1.5× to 17× depending on the application and hardware comparison (single high-end GPU vs. multi-core CPU nodes). Examples cited in release coverage include strong CPU-core equivalence numbers (often 150–250+ cores per GPU in multiphase or particle-heavy cases) along with lower energy consumption.
New workstation GPU support includes the NVIDIA RTX PRO 6000 Blackwell and AMD Radeon AI PRO R9700. Multi-user GPU visibility controls and improved sparse matrix handling for the coupled flow solver further improve practical usability on shared systems.
Energy consumption monitoring and reporting is also available, allowing teams to track power use by simulation phase for sustainability and cost analysis.
Electromagnetics: New Finite Element Magnetic Field Solver
A major multiphysics highlight is the new Finite Element (FE) Magnetic Field solver(Phase 1), designed for full 3D e-machine simulation. It targets high-fidelity magnetic field, torque, loss, and field predictions while coupling more efficiently with thermal and fluid solvers inside the same STAR-CCM+ environment. Siemens reports substantial performance gains over earlier magnetic vector potential approaches (typically ~10×–20× for industrial-grade EV traction motors with hairpin windings, depending on the comparison baseline and core count). An EMAG-Thermal template and improved connectivity with Simcenter E-Machine Design (including better handling of detailed windings) aim to reduce the friction of multiphysics e-motor workflows. Supporting improvements include better temperature-dependent nonlinear material handling and extended electrostatic force density options.
Multiphase, Particles, and Combustion
Beyond pure GPU enablement, multiphase and reacting-flow capabilities continue to mature:
Expanded GPU support for complex chemistry and EBU models accelerates detailed combustion and emissions work (relevant to turbomachinery and industrial combustors).
Lagrangian multiphase Phase 1 delivers faster particle tracking on GPU.
Fluid Film / DMP and EMP gains improve vehicle water management, fluidized beds, sprays, and related applications.
Additional model refinements (drag options, interpolation controls, porous media compatibility with phase change, etc.) increase robustness and flexibility.
GPU acceleration of the Lagrangian spray solver specifically benefits liquid fuel injection studies in turbomachinery, helping reduce solve times from days toward hours in demanding cases.
Smoothed Particle Hydrodynamics (SPH) Advances SPH continues its integration path:
Conjugate heat transfer (direct coupling with the energy solver for fluid-solid temperature resolution).
Aerodynamic forcing (mapping of FV flow fields onto SPH with drag coupling for airflow–liquid interactions).
Temperature-dependent viscosity, improved inlets (volume fraction control), translation motion support, and better data analysis/particle refinement options.
These make SPH more practical for applications involving lubrication, gear systems, free-surface flows with heat transfer, or combined air–liquid effects.
Acoustics
A native Frequency Solver provides an all-in-one frequency-domain capability for pure acoustics and duct/HVAC noise problems, reducing the need to transfer data between tools. UI refinements clarify noise source weighting and acoustic damping terminology.
Meshing Productivity
Distributed-memory parallel Surface Remesher (Phase 1) scales to 16 CPU cores and delivers up to ~2.1× faster surface remeshing on complex models while preserving quality and face count.
Automatic weak contact creation within Boolean operations (unite/intersect/subtract), reducing reliance on imprint steps.
Additional meshing conveniences (including cylindrical templates for trimmed cell meshing in annular/cylindrical geometries) improve flow-aligned grids and convergence behavior.
Post-Processing, Visualization, and Comparison
Comparative Data Analysis lets users accumulate and compare results from multiple simulations (via mapped field functions and solution histories) inside a single file, supporting side-by-side cause-and-effect investigation of different physics or solver setups.
Studio Scene enhancements include Deep Learning Super Sampling (DLSS) for higher frame-rate photorealistic rendering, HDR environment maps, and export to Universal Scene Description (.usdz) for Omniverse and mixed-reality collaboration.
Augmented Reality / Mixed Reality passthrough with hand tracking (Web Viewer + compatible headsets such as Meta Quest 3) allows projection of CFD results onto physical prototypes.
The previous STAR-CCM+ Viewer is replaced by a downloadable Web Viewer-based desktop application (Simcenter Viewer) with better large-file handling.
AI and Design Exploration
Multi-GPU support for the PhysicsAI add-on enables significantly faster training (reported ~3× when scaling from 1 to 4 GPUs in cited benchmarks, contributing to overall “up to 3× faster” AI model training claims). This helps teams iterate surrogate models more rapidly for design exploration.
Platform and Other Notes
The unified server/client runtime introduced earlier is now the default. Various smaller usability, data-analysis, and interoperability refinements round out the release. Note that certain legacy items (e.g., the old standalone Viewer and Block-Mapped Coordinate Systems) reach end-of-support as previously announced.
Summary
STAR-CCM+ 2606 strengthens the product’s position for high-fidelity, multiphysics, GPU-accelerated simulation—particularly for electrified powertrains (EMAG + thermal + fluid), multiphase and particle-laden flows, combustion, and free-surface/SPH applications—while improving day-to-day productivity through faster meshing, better result comparison, and modern visualization/AR capabilities.
Engineers working on electric machines, advanced air mobility, turbomachinery, vehicle water management, or any GPU-friendly multiphase/combustion problems should see the most immediate impact. As always, check the official Siemens release notes, feature list, and Support Center documentation for the complete set of changes, supported platforms, and any migration considerations.
Accelerate liquid fuel injection simulations with GPU computing
.png?upscale=true&width=1120&upscale=true&name=industry%20newsletter%20footer%20%20(2).png)
