ThinkSystem SD530 and D2 EnclosureDatasheet
Ready to adapt when you are
Ready For What’s Next
The Lenovo ThinkSystem SD530 is a single platform designed to excel not only at critical enterprise workload environments (such as virtualization, hyperconverged infrastructure, and cloud), but also at High Performance Computing (HPC) and Artificial Intelligence (AI). Combining the efficiency and density of blades with the value and simplicity of rack-based servers, SD530 could very well be the most adaptable server on Earth.
ThinkSystem SD530 consists of a modular 2U Lenovo D2 Enclosure containing up to four front-access SD530 servers (nodes). Each node, incorporating two second-generation Intel® Xeon® Processor Scalable family CPUs, achieves up to a 36% performance improvement compared to the previous generation.¥
The innovative design of the D2 enclosure provides the flexibility to meet varied needs in the modern data center. For example, multiple D2 Enclosures can be easily daisy-chained together and then managed as a single unit, which reduces cabling costs by as much as 92%, compared to the previous generation,* and simplifies administration.
Today’s IT managers are always driven to do more with less. The SD530 handles twice the workloads per U of traditional 1U servers. In a single 42U rack you can mount up to 76 servers with: Up to 152 processors, 4,256 cores, 77.8TB of memory, and 3.6PB of storage.† SD530 provides 32 more cores per U than the previous generation* while maintaining drive density of up to 6 SFF drives per node with up to two direct-connect NVMe SSDs—all within industry-standard racks.
For software-defined storage and hyperconverged workloads that need significant local storage, SD530 offers ideal storage density and capacity. Support for U.2 SSDs provides ample performance to meet increasing business demands. Internal M.2 boot drives enable increased capacity and significantly more reliability than the SATADOM solutions used today.
Maximum Compute Capability
SD530 is engineered to run the highest-core Intel® Xeon® Platinum processors to power through your most demanding HPC or AI workloads. With the increasing demand for GPU technology across many use cases including VDI, HPC, and Machine Learning, the SD530 supports a variety of GPUs—including the latest NVIDIA Tesla V100. The innovative D2 Enclosure accommodates an optional 1U tray that supports two GPUs or accelerators per node (two trays maximum per D2 Enclosure).
Additional memory and storage technology enhancements provide exceptional performance and flexibility in any environment:
- 50% increase in memory channels and capacity, with an 11% increase in memory speed vs. the prior generation 2U/4-node platform*
- Twice the SAS bandwidth of the previous generation, with support for 12Gb SAS
- Twice the storage capacity of the prior platform* for hyperconverged environments
The SD530 delivers the density of blades with the economics of rack systems.
By utilizing a single platform designed to excel at both enterprise and HPC workloads, ThinkSystem SD530 was engineered for flexibility and enables you to reduce qualification and test times and increase your flexibility by using common components and management.
This is accomplished through the revolutionary design of the D2 Enclosure. The front of the unit holds up to four nodes that are “warm-swappable” to maximize uptime. Individual nodes can be removed without powering down the remaining three nodes. Each SD530 node contains processors, memory, and up to six 2.5" storage devices.
The rear of the D2 Enclosure contains a “shuttle” with power supplies, fans, adapter slots, and networking and management ports. By disaggregating the adapter slots and NICs (both LOM and PCIe) from the nodes, the unique shuttle designs provide maximum I/O flexibility.
The enclosure supports either of two shuttles:
- The x16 Simple-Swap PCIe Shuttle, with four PCIe x16 low-profile adapter slots (one per node)—maximizing I/O performance
- The x8 PCIe Shuttle, with eight PCIe x8 low-profile adapter slots (two per node)—maximizing the number of adapter slots
Both shuttles support 10Gb Base-T, 10Gb SFP+, or no-LOM options, enabling you to install only the LOM/NICs you need for the best networking options within your budget. The shuttles support a wide array of high-speed interconnect fabrics, including OmniPath and InfiniBand for HPC/AI environments, as well as a full range of Ethernet and Fibre Channel cards for more traditional enterprise and cloud environments.
Modular Daisy Chain Management
Besides the D2 enclosure, Lenovo also provides a modular enclosure for daisy chain management. The modular enclosure provides lower ToR ports cost, lower cable costs, and less rack complexities.
The revolutionary design of rear shuttles used with the D2 Enclosure provides maximum I/O flexibility for the SD530 ultra-dense platform.
Lenovo XClarity Controller is an all-new hardware embedded management engine common in every ThinkSystem server. XClarity Controller features an uncluttered graphical user interface, industry standard Redfish-compliant REST APIs, and enables booting in half the time of prior generation servers, with up to 6x faster firmware updates.
The D2 shuttle contains a Scalable Management Module (SMM) that controls the fans and power as well as combines the XClarity Controller of each node into a single port. The optional dual-port SMM allows daisy chaining of enclosures reducing cabling cost and complexity by as much as 92%, compared to the previous generation.**
Lenovo XClarity Administrator is a virtualized application that centrally manages ThinkSystem servers, storage, and networking. Via reusable patterns and policies, it ramps up and scales infrastructure provisioning and maintenance. It serves as a central integration point to extend your data center management processes to physical IT. Running XClarity Integrators in external IT applications, or integrating through REST APIs, helps you further speed services provisioning, streamline IT management, and contain costs.
Lenovo servers continue to be the industry’s #1 most reliable§, with the industry’s highest customer satisfaction‡ rating.
Lenovo Services and Security
Supporting the full life cycle of your investment, Lenovo Services are the perfect complement to Lenovo’s world-class enterprise products: Lenovo ThinkSystem servers, storage and networking offerings.
As the trusted services provider for thousands of companies around the world, Lenovo has the expertise and experience to help with everything from solutions architecture to implementation, integration, and migration, to proactive management services.
Lenovo Services guarantee you a superior service experience, and are delivered through Lenovo service professionals and the network of Lenovo Authorized Service Providers.
Lenovo (HKSE: 992) (ADR: LNVGY) is a US$45 billion Fortune 500 company and a global technology leader in driving Intelligent Transformation. Lenovo’s data center solutions (ThinkSystem, ThinkAgile) are creating the capacity and computing power that are changing business and society.
|2U rack enclosure; 4 independent compute nodes|
|Processor||Up to 2 second-generation Intel® Xeon® Platinum processors, up to 205W|
|Memory||Up to 2TB in 16x slots, using 128GB DIMMs; 2933MHz TruDDR4|
|Expansion Slots||1x shuttle per D2 enclosure: either x8 PCIe Shuttle with 8x PCIe 3.0 x8 slots (2 per node); or x16 PCIe Shuttle with 4x PCIe 3.0 x16 slots (1 per node). Up to 2x external 1U trays (with up to 2 GPUs per node each)|
|Drive Bays||Up to 24x (6x per node) hot-swap 2.5" SAS/SATA HDDs/SSDs; up to 16x (4x per node) hot-swap 2.5" NVMe SSDs|
|Network Interface||Optional 8-port EIOM 10Gb SFP+ (2 ports per node); optional 8-port EIOM 10GbaseT RJ45 (2 ports per node)|
|Power Supply||2x hot-swap/redundant 1+1 1600W/2000W; or 2x non-redundant 1100W|
|Hot-Swap Components||Power supplies, fans, SAS/SATA/NVMe storage devices; compute nodes are warm-swap|
|RAID Support||SW RAID supports JBOD; Entry RAID; optional HW RAID 12Gb supports JBOD; plus M.2 boot support with optional RAID|
|Systems Management||XClarity Controller embedded management, XClarity Administrator centralized infrastructure delivery, XClarity Integrator plugins, and XClarity Energy Manager centralized server power|
|GPU Tray Support||The GPU Tray supports one or two GPU cards; Maximum of two GPU trays per D2 Enclosure|
|OSes Supported||Microsoft, Red Hat, SUSE, VMware. Visit lenovopress.com/osig for details.|
|Limited Warranty||3-year customer replaceable unit and onsite service, next business day 9x5, optional service upgrades|
¥ Based on Intel internal testing, August 2018. † Assumes 4U of rack occupied by networking or other components. * Compared to Lenovo NeXtScale nx360 M5. § ITIC 2018 Global Reliability Report. ‡ TBR x86-based Servers Report 2018. ** Lenovo recommends limiting the number of modular enclosures in a daisy chain to 7 modular enclosures.
Related product families
Product families related to this document are the following:
© 2019 Lenovo. All rights reserved.
Availability: Offers, prices, specifications and availability may change without notice. Lenovo is not responsible for photographic or typographic errors. Warranty: For a copy of applicable warranties, write to: Lenovo Warranty Information, 1009 Think Place, Morrisville, NC, 27560. Lenovo makes no representation or warranty regarding third-party products or services.
Trademarks: Lenovo, the Lenovo logo, Lenovo Services, Lenovo XClarity, NeXtScale, ThinkAgile, ThinkSystem, and TruDDR4 are trademarks or registered trademarks of Lenovo. Intel® and Xeon® are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. Microsoft® is a trademark of Microsoft Corporation in the United States, other countries, or both. Other company, product, or service names may be trademarks or service marks of others.
Document number DS0003, published November 6, 2017. For the latest version, go to lenovopress.com/ds0003.