SCSK

Solver
ADVENTURECluster has developed a unique solution method (*CGCG method, a hybrid solution method with the direct method), which enables fast calculations for a variety of analyses.

About ADVENTURECluster's solver CGCG method

Unique solution combining region segmentation and iterative method
High parallel performance and reduced memory consumption

In general numerical programs using the finite element method, the decline in parallel efficiency becomes a major problem as the number of parallels increases. In addition, as the number of nodes increases, the larger the matrix to be solved becomes, the more memory is required, requiring a high-specification computer. However, ADVENTURECluster has developed a unique solution (*CGCG method, a hybrid solution method with the direct method) that maintains high parallel efficiency and reduces memory consumption even when the number of parallels increases.
※CGCG法（Coarse Grid Based Conjugate Gradient）

General application ADVENTURECluster
No. of parts 1～10 parts 1～500 parts
No. of nodes About 1 million max. About 30 million max.
*Achieved;200 million nodes
ADVC

Solver Features

Feature 1 : High-speed calculations

Ultra-fast calculation possible

One feature of ADVENTURECluster is its fast computation time. Shorter computation time increases the number of CAE studies within a limited time, which sometimes makes it difficult to perform sufficient evaluation studies within a limited time. ADVENTURECluster uses a highly convergent algorithm to complete calculations in a short time. By shortening the solver calculation time, the number of evaluation studies within a limited time can be increased. By completing calculations in a shorter time, more design alternatives can be evaluated. This improves design quality and development speed.

ADVC

Feature2 : Parallel efficiency

By increasing the number of parallels, time-consuming analysis can be completed in a shorter time.

The usual finite element method software has a major problem of parallel efficiency decline as the number of parallels increases. ADVENTURECluster has developed its own algorithm CGCG method, so parallel efficiency does not drop.

ADVC
ADVC

The graph on the left is an overall engine block analysis consisting of approximately 6 million elements and 39 parts. The number of parallelism is taken on the horizontal axis and the computation time on the vertical axis, and the growth rate of the ADVENTURECluster computation is expressed in terms of parallelization efficiency.

Feature3 : Memory saving

Even if the analysis scale is large, it can be analyzed with less memory!

Finite element method software requires more memory as the scale of analysis increases, requiring an expensive computing environment. ADVENTURECluster requires very little memory to perform calculations, making it possible to perform larger-scale analyses than ever before even in your current computing environment. In addition, ADVENTURECluster's low memory requirements are also advantageous when running calculations and pre/post-processing on the same machine.

ADVC

The above figure shows a series of analysis flow in the automotive engine reliability analysis assembly process. Even a model with approximately 6 million elements can be analyzed with as little as 39,8 GB of memory.

ADVC

List of Analysis Functions

List of ADVENTURECluster analysis functions

ADVENTURECluster Various analysis functions
Analysis type Static analysis Static analysis Elastic analysis
Contact analysis
Large deformation analysis
Elasto-plastic analysis
Creep and viscoelastic analysis
Hyper-elastic analysis
Rate-dependent non-linear kinematic hardening analysis (Material model suitable for high-temperature metallic materials)
Lead-free solder creep analysis
Dynamic analysis Dynamic analysis Transient response analysis (including direct time integration and explicit dynamic analysis)
Eigenvalue analysis
Modal frequency response analysis
Random response analysis
Fatigue analysis Fatigue analysis Low-cycle fatigue analysis
High-cycle fatigue analysis
Fatigue crack growth analysis
Fracture mechanics analysis Stress intensity factor
J-integral
Optimization analysis Optimization analysis Topology optimization analysis
Shape optimization analysis
Heat transfer analysis Heat transfer analysis Steady-state heat transfer analysis
Non-steady heat transfer analysis
Coupling analysis Coupled structural and heat transfer analysis
CAD Interfaces CATIA Interface
Creo Interface
NX Interface
Parasolid Interface
STEP, IGES Interface
Collabo Interfaces SCRYU/Tetra Interface

CONTACT

Contact Us & Request Info

TEL : +81-3-5859-3012

Office Hours : 9:30～18:00（Excluding Saturdays, Sundays, and holidays）