Simulation and modeling applications, such as computer-aided engineering (CAE) and computational fluid dynamics (CFD), are designed to run on scalable, high-performance clusters. To support those applications at scale, modern high-performance computing (HPC) systems require multi-core processors, high-bandwidth fabrics, and broad input/output (I/O) capabilities.
Because of the complexity and variety of technologies available on the market, assembling an HPC system can be time-consuming, requiring additional effort for research, evaluation, and deployment. In addition, integration and configuration of selected components can impact the performance of the solution. The level of expertise required to properly configure the combined solution can intimidate small and medium organizations. Even larger enterprises considering the benefits of HPC clusters must weigh the time and effort it takes to ramp up capabilities. These barriers slow down adoption of HPC clusters even though the return on the investment (ROI) can be significant.
MEGWARE HPC Simulation and Modeling Solution accelerates the process of defining and deploying an HPC cluster for these advanced applications. Benefits include:
- Performance-optimized for highly scalable simulation and modeling applications that use Message Passing Interface (MPI) libraries.
- Reduces the time required to evaluate, select, purchase, deploy, configure, and support a workload-appropriate solution.
- Improves delivery and uptime through simplified deployment and advanced diagnostics tools.
- Compatible with other HPC applications listed in Intel’s Intel® Scalable System Framework (Intel® SSF) application catalog.
MEGWARE engineers plans, designs, delivers and installs entire cluster solutions - racks, servers, PDUs, network equipment, storage as well as software. MEGWARE brings long-standing expertise to customers from academia to industrial clients who want to have a full-service package, including premium level support over multiple years.
In addition to design, building, and installing custom systems, MEGWARE offers its own hardware and software components for optimized HPC performance and management, such as MEGWARE ColdCon® cooling technology, SlideSX® server chassis, and ClustWare® management software, plus other products. These technologies can complement the overall HPC cluster management and operation.
The MEGWARE HPC Simulation and Modeling Solution comprises several key hardware and software components: Intel® Xeon® Scalable processors, the Intel® Omni-Path Architecture (Intel® OPA) HPC fabric, and other Intel® technologies that benefit HPC applications. Component choices deliver optimized performance for MPI-based simulation and modeling applications. Supportability is also addressed with the inclusion of Intel® Cluster Checker, which provides expert systems advice for administrators to use in keeping a cluster functioning.
MEGWARE MiriQuid® Compute Platform available as air-cooled 2U 4-node system or 6U 10-node direct liquid-cooled system, based on latest MEGWARE ColdCon® technology.
|INGREDIENT||INTEL® SELECT SOLUTIONS FOR SIMULATION AND MODELING CLUSTER CONFIGURATION DETAILS|
|PLATFORM||MEGWARE MiriQuid® Compute Node|
|PROCESSOR||2 x Intel® Xeon® Gold 6148 processors at 2.40 GHz, 20 cores/40 threads|
|MEMORY||192 GB (12 x 16 GB 2666 MHz DDR4)|
4 GB memory per processor core and all memory channels populated
|LOCAL STORAGE||256 GB Enterprise SSD|
|SOFTWARE||Linux operating system (RHEL, CentOS, SLES)|
Intel® Parallel Studio XE 2018 Cluster Edition
Intel® Cluster Checker 2018
|COOLING||1. Standard air cooling|
2. TCO-optimized with MEGWARE's extremely efficient ColdCon® direct liquid cooling technology, supporting high coolant inlet temperatures of up to 50℃