Blade server RSC Tornado TDN531
- Liquid cooled compact blade server with two 2nd Gen Intel Xeon Scalable processors.
- Up to 4 NVMe solid state drives
- Can be used for high performance computing or server for data center.
- Hardware management controller inside.
- Compatible with RSC Tornado racks, needs one installation slot.
- Power redundancy N+3 with power redundant module.
More solid state drives in a high-dense blade server
- Two NVMe disks with new form factor — EDSFF E1.S «short rooler» with increased capacity and performance.
- Two NVMe диска with widely used form factor M.2 — flexiblility of choice and availability of SSDs with Intel® Optane™ technology.
- Fast storage — up to 15.6 TB inside a blade server.
- Plus SATA solid state drive for operating systems files.
1 TiB of RAM with Intel® Memory Drive Technology (IMDT)
When combined with Intel® Optane™ SSDs, Intel® Memory Drive Technology transparently integrates the drive into the memory subsystem and makes it appear like DRAM to the operating system and applications.
2nd Gen Intel® Xeon® Scalable
- Higher Per-Core Performance: up to 28 cores, delivering high-performance and scalability for compute-intensive workloads across compute, storage, and network usages.
- Intel® Deep Learning Boost (Intel® DL Boost) with Vector Neural Network Instruction (VNNI) bring enhanced artificial intelligence inference performance, with up to 30X performance improvement over the previous generation, 2nd Gen Intel® Xeon® Scalable processors help to deliver AI readiness across the data center, to the edge and back.
- Greater Memory Bandwidth: 50 percent increased memory bandwidth and capacity. Support for six memory channels per socket, with speeds up-to 2933 MT/s (1 DPC).
- Intel® Advanced Vector Extensions 512 (Intel® AVX-512): With double the FLOPS per clock cycle compared to previous-generation Intel® Advanced Vector Extensions 2 (Intel® AVX2), Intel® AVX-512 boosts performance, and throughput for the most demanding computational tasks in applications, such as modeling and simulation, data analytics and machine learning, data compression, visualization, and digital content creation
Reliable cooling for hottest processors
Even CPUs with TDP 205 W work stably and effectively. No throttling and losing performance!
Node Management Controller (NMC)
- Installed into each blade server for integration into a single composable architecture.
- Performs advanced monitoring, management, monitors the prevention of possible emergencies and can also perform the tasks of the RSС BasIS Orchestration Platform, providing its resources for building composable computing and storage systems.
Hi-speed interconnect up to 100 Gbit/s
Using high performance network technologies like Intel Omni-Path, Mellanox Infiniband EDR/HDR100, 100Gb/s Ethernet connected through PCI Express 3.0 x16.
High-density hyperconverged system
In one RSC Tornado TCC153B rack, 800 x 800 x 2000 mm:
- 153 blade servers RSC Tornado TDN531;
- 306 Intel Xeon Scalable CPUs;
- 8 568 processor cores;
- up to 2 PB of high-speed software defined storage.
100% liquid cooling
- no air cooling required;
- high reliability – no fans and mechanically moving parts inside the servers and rack;
- "hot water" cooling – free cooling all year round, PUE 1,04.
Using accelerators
- Server blade TDN531 supports the installation of up to two of any accelerators connected to PCI Express x16 ports, for example, such as NVIDIA A100, NVIDIA Tesla V100, Intel FPGA, AMD Radeon.
- Accelerators are completely liquid-cooled using a cooling plate; no air cooling is required.
- With accelerators installed, the server blade occupies two slots in a rack.
Easy to use with RSC BasIS management platform
- Cluster installation complexity: By providing proven and complete set of HPC Cloud configuration templates and instruments.
- Creating Software Defined Storage: based on software defined configurations, fully optimized for the user needs, with storage elements provisioned dynamically on request and exported either as a block or a filesystem.
- Data center management complexity: By using a single approach to manage all data center layers: from infrastructure to user workload.
- Data center knowledge management: By providing powerful APIs for data gathering and analysis, supported by extendable visualisation tools.
- Data center Operation Automatization: By allowing system operators to create flexible automation scenarios based on collected data.
- Platform Extensibility: Advanced Platform Framework and API to develop custom components based on micro-agent architecture
