Home Services Products About Brands Logistics Contact
The Lenovo ThinkSystem server family, tower, rack and dense multi-node systems
Lenovo Business Partner
Lenovo ThinkSystem

Lenovo ThinkSystem servers and storage, specified and delivered.

Celltronics is a Lenovo Business Partner. We quote, configure, and deliver ThinkSystem rack, tower and dense servers, GPU platforms, and ThinkSystem DG, DM and DE storage arrays.

Why Lenovo, Why Celltronics

Two generations run side by side, and that is where buyers get caught.

Lenovo's V4 platforms did not replace the whole V3 line. Several V3 systems remain the current answer for higher socket counts and GPU-dense builds, while V4 leads on mainstream rack compute. Knowing which generation applies to your requirement is most of the work, and it is the part a spec sheet will not tell you.

Sourced through TD Synnex

Established distribution relationships for Lenovo ISG, which is what turns a configuration into a real quote with a real lead time.

Quote and lead time up front

You get a complete bill of materials with pricing and realistic availability, so procurement and project planning line up.

International fulfillment

Domestic and international shipping through established freight forwarders, including consolidated and palletized export deliveries.

The Portfolio We Supply

ThinkSystem servers, GPU platforms, and DG, DM and DE storage.

These are the Lenovo lines Celltronics quotes and supplies. Platform availability and exact configurations change as Lenovo refreshes its generations, so treat the details below as a starting point and ask us to confirm current options for your build.

ThinkSystem rack and tower servers

The V4 generation moved the mainstream rack line to Intel Xeon 6. The V3 platforms remain current for higher socket counts and for AMD based configurations.

Mainstream rack, V4

ThinkSystem SR630 V4 and SR650 V4

  • SR630 V4, 1U, two sockets, Intel Xeon 6 with P-cores or E-cores, with GPU support for acceleration
  • SR650 V4, 2U, two sockets, Intel Xeon 6 with P-cores
  • The E-core option on the SR630 V4 suits scale-out work where core count per watt matters more than per core speed
Four socket and mission critical

ThinkSystem SR860 V4 and SR850 V4

  • SR860 V4, 4U, two or four sockets, Intel Xeon 6700 series with P-cores
  • Up to 86 cores and 172 threads per system
  • Up to 16TB of memory across 64 DDR5 DIMMs at 6400 MHz, or CXL 2.0 modules in E3.S 2T form factor
  • Drive options up to 48 x 2.5 inch, or 32 x E3.S 1T NVMe plus 24 x 2.5 inch
  • Also supports up to 8 single-wide 75W GPUs or 4 double-wide 400W GPUs
V3 generation, still current

SR950 V3, SR860 V3, SR665 V3 and SR655 V3

  • SR950 V3, eight socket, for the largest scale-up workloads
  • SR860 V3, four socket
  • SR665 V3 and SR655 V3, two socket, GPU capable
  • SR645 V3 and SR635 V3, single socket
Tower

ThinkSystem ST650 V3, ST250 V3, ST50 V3

  • ST650 V3, two socket tower
  • ST250 V3, ST50 V3 and ST45 V3, single socket
  • For offices and branch sites without a rack, or where noise and footprint govern the choice

GPU and AI systems

Lenovo splits accelerated compute between GPU capable mainstream servers and dedicated GPU platforms. Which side of that line you need depends on GPU count, not on workload label.

GPU capable, 2U

ThinkSystem SR650a V4

  • 2U, two sockets, Intel Xeon 6 with P-cores
  • Support for four double-wide GPUs, including NVIDIA H100 with NVLink connectivity
  • The practical choice when you need real GPU capacity without moving to a dedicated AI chassis
Dedicated GPU platform

ThinkSystem SR680a V4

  • Lenovo's dedicated GPU server for AI and HPC workloads
  • Positioned for maximum acceleration rather than mixed use
  • Ask us to confirm the current accelerator options and cooling method for your configuration
V3 GPU-dense platforms

SR675 V3, SR680a V3 and SR685a V3

  • Two socket GPU-dense platforms, still current alongside the V4 line
  • Used where a proven V3 configuration is already qualified in your environment
Dense and multi-node

ThinkSystem SC750 V4, SC777 V4 and the SD series

  • SC750 V4 and SC777 V4, the V4 dense compute platforms
  • SD665-N V3, SD650-N V3, SD550 V3, SD535 V3 and SD530 V3 multi-node systems
  • For HPC clusters where nodes per rack and interconnect drive the design

ThinkSystem storage, DG, DM and DE

Three series that are genuinely different, not three price points for the same thing. Picking the wrong one is the most common storage mistake we see.

Performance, all NVMe

ThinkSystem DM series

  • DM3200F, DM5200F and DM7200F, end-to-end NVMe, 2U, two controllers
  • DM5200H for hybrid flash configurations
  • Unified storage, for medium enterprise workloads where latency is the constraint
Capacity, all QLC flash

ThinkSystem DG series

  • DG5200 and DG7200, all-QLC flash, 2U, two controllers
  • Capacity optimized, for storage efficiency without dropping back to spinning media
  • The right answer when you want flash economics across a large dataset
Mid-range and entry

ThinkSystem DE series

  • DE4800F all-flash and DE4800H hybrid, mid-range
  • DE4200H hybrid, entry level
  • Positioned for analytics, video surveillance, and backup targets
Choosing between them

DM, DG or DE

  • DM when latency governs, DG when cost per terabyte of flash governs
  • DE when the workload is capacity and throughput rather than latency
  • We size the array against the actual IO profile, not the headline capacity
What We Build For

The workloads behind the hardware.

Tell us the workload and we will work backward to the platform. These are the requirements we see most often.

Virtualization and Private Cloud

Consolidated compute with the memory and storage headroom to keep density high without starving the guests.

Databases and ERP

Two and four socket platforms for transactional systems, with memory footprints up to 16TB where the dataset needs to sit in RAM.

AI Training and Inference

GPU capable mainstream servers where four accelerators are enough, and dedicated GPU platforms where they are not.

HPC and Technical Computing

Dense multi-node systems for simulation and modeling, sized for nodes per rack and the interconnect.

Primary and Secondary Storage

DM for latency sensitive primary storage, DG for flash economics at capacity, DE for analytics, surveillance and backup targets.

Branch and Distributed Sites

Tower and single socket rack platforms for offices and remote sites, specified so a later move into a rack is not a rebuild.

How We Work

From requirement to rack.

Requirement

We start with the workload, the node count, the power and space you have, and the date you need it running.

Configuration

We build the bill of materials against Lenovo's current V4 and V3 platforms and flag anything that will bottleneck in practice.

Quote

Pricing and realistic lead times, with alternatives where a generation change or a different chassis saves money or time.

Delivery

Order placement, tracking, and shipping to your site or forwarder, domestic or international.

Get a Configuration and a Price

Tell us the workload. We will spec the system.

Send the requirement to info@celltronics.com or call 954-885-2600. Our team is here 24/7.

Request a Lenovo Quote