| Marketing Information | Unprecedented Performance
The Lenovo ThinkSystem SR665 utilizes AMD EPYC™ architecture to deliver the next era of solution performance and TCO optimization for softwaredefined workloads, database, big data & analytics, virtualization, virtualized desktop infrastructure (VDI), and HPC/AI workloads.
The combination of two AMD EPYC 7003 Series CPUs with class-leading memory speed, storage, and GPU density rapidly outpace the power of prior generation two-socket servers. Lenovo's lauded system reliability, management capabilities, and security infrastructure layer on to the exceptional value that the ThinkSystem SR665 brings to the data center.
The modern data center demands ever-expanding capability, requiring server technology to deliver more performance that tests the limits of physics. With enterprise-class AMD EPYC™ 7003 Series processors, the new standard for the modern data center CPU, the ThinkSystem SR665 features an unprecedented 128 cores and 128 Gen4 PCIe lanes in a server to maximize server performance and decrease network bottlenecks critical for HPC/AI workloads.
Adaptive Design
The ThinkSystem SR665 is an all-purpose workhorse, adaptable to numerous enterprise workloads to address a wide range of data management solutions in hybrid data center environments.
The dual AMD EPYC 7003™ Series CPUs combined with support for multiple GPUs streamlines performance for virtual desktop infrastructure (VDI) instances, providing the technical resources to aid in scale-out VDI instances for a growing enterprise workforce.
Cutting-edge storage configurations include choices ranging from high capacity HDDs to lightning-quick NVMe drives to maximize both data capacity and throughput performance. These storage features render the system ideal for database and big data/ analytics workloads.
Outfitted with up to 8x PCIe 4.0 slots, the ThinkSystem SR665 provides the expandability, networking scalability, and speed necessary to ensure that I/Ointensive workloads such as software defined networking, and grid computing do not encounter bottlenecks that reduce full enterprise efficiency. |
