Accelerating artificial intelligence performance The SR670 delivers scalable performance for HPC and AI workloads demanding best-of-breed GPU acceleration.

Accelerating artificial intelligence performance

The SR670 delivers scalable performance for HPC and AI workloads demanding best-of-breed GPU acceleration.

DESCRIPTION

SPECIFICATIONS

Form Factor

Full-width 2U enclosure

Processors

2x second-generation Intel® Xeon® Scalable Processors (up to 205W) per node

Memory

Up to 1.5TB using 24x 64GB 2933MHz TruDDR4 3DS RDIMMs per node

I/O Expansion

Up to 3 PCIe adapters: 2x PCIe 3.0 x16 + 1x PCIe 3.0 x4 slots

Acceleration

Up to 4 double-wide, full-height, full-length GPUs (each PCIe 3.0 x16 slots), or up to 8 single-wide, fullheight, half-length GPUs (each PCIe 3.0 x8 slots)

Management Network Interface

1x RJ-45 for dedicated 1GbE system management

Internal Storage

Up to 8x 2.5″ hot-swap SSD or HDD SATA drives in rear bays

Up to 2x non-hot-swap M.2 SSDs, 6Gbps SATA in internal bays

RAID Support

SW RAID standard; optional HBA or HW RAID with flash cache

Power Management

Rack-level power capping and management via Extreme Cloud Administration Toolkit (xCAT)

Systems Management

Remote management using Lenovo XClarity Controller; 1Gb dedicated management NIC

OS Support

Red Hat Enterprise Linux 7.5

Limited Warranty

3-year customer replaceable unit and onsite limited warranty, next business day 9×5, service upgrades available