Lenovo 4XG7A38082 Thinksystem SR590/SR650 Intel Xeon Gold 6226R 16C 150W 2.9GHZ Processor Option

Lenovo 4XG7A38082 Intel Xeon Gold 6226R 16-Core Processor Option Kit

Overview

The Lenovo 4XG7A38082 is a factory-new processor option kit delivering Intel's 2nd Generation Xeon Scalable architecture to the ThinkSystem SR590 and SR650 server platforms. Built around the Intel Xeon Gold 6226R — a 16-core, 32-thread Cascade Lake processor running at 2.9 GHz base with 3.9 GHz Turbo Boost — this kit is purpose-matched for compute-dense workloads where single-socket or dual-socket server builds need a credible step up in throughput without jumping to the higher TDP bins in the Gold or Platinum family. The 150W TDP sits in a deliberate middle range: more headroom than the entry Gold SKUs, while still fitting standard server cooling configurations designed for the SR590 and SR650 chassis.

Delivered as a tray-packaged component through commercial channels, the 4XG7A38082 is the correct part number when configuring or expanding a ThinkSystem SR590 or SR650 with a second socket or replacing a processor at matching performance tier. It is not a retail-boxed CPU — it ships without a cooler, which is expected and correct for server platforms where the chassis heatsink is specified separately.

Key Features

• 16 Cores / 32 Threads on Cascade Lake: Intel's Cascade Lake microarchitecture brings hardware-level mitigations for speculative-execution vulnerabilities baked in — critical for server deployments that cannot absorb the OS-level performance penalty of software mitigations. For workloads running virtualized VMS servers, analytics engines, or database back-ends, the 32 hardware threads keep context-switch overhead low across concurrent task loads.
• 2.9 GHz Base / 3.9 GHz Turbo Boost: The 1 GHz spread between base and boost is meaningful for mixed workloads. Latency-sensitive single-threaded processes (licensing daemons, certain VMS event triggers) will opportunistically reach 3.9 GHz, while sustained multi-threaded tasks (video transcoding, analytics inference) stabilize at base. Plan thermal and power budgets around base, not boost.
• 22 MB L3 Cache: At 22 MB, the last-level cache provides roughly 1.375 MB per core — adequate for analytics pipelines that benefit from keeping working datasets close to execution units. For workloads with large streaming data (continuous video decode across many channels), cache size matters less than memory bandwidth, which the dual UPI links address at the platform level.
• Dual UPI Links at LGA 3647 (Socket P): Two Ultra Path Interconnect links provide the CPU-to-CPU fabric in dual-socket SR650 configurations. This is the spec that determines whether a two-socket build actually scales: dual UPI means both sockets share coherent memory access across the full 2 TB addressable space (2x 1 TB per socket), rather than hitting a single-link bottleneck on NUMA-intensive workloads.
• 1 TB Maximum Memory Per Socket: Each socket can address up to 1 TB of RAM using supported DDR4 LRDIMM or RDIMM configurations. For NVR workloads, AI inference servers, or virtualization hosts running many concurrent VMs, this ceiling is rarely a constraint — but knowing it exists matters when planning memory expansion for high-channel-count video analytics deployments.
• 14 nm Lithography: Cascade Lake on 14 nm is a mature, well-characterized node. Thermal behavior is predictable, and the process has years of field validation behind it. For data centers where server stability and predictable power draw matter more than cutting-edge node shrinks, this is a practical advantage over first-generation parts.
• 64-Bit Operating Mode: Full 64-bit instruction set execution is table stakes for any modern server OS and workload, but worth confirming when sourcing processor upgrades — this part fully supports x86-64, AVX-512, and the instruction extensions expected by current enterprise software stacks including VMware ESXi, Windows Server, and major Linux distributions.

Integration & Compatibility

The 4XG7A38082 is explicitly qualified for two ThinkSystem platforms: the SR590 (2-socket 1U) and the SR650 (2-socket 2U). Both use the LGA 3647 Socket P interface. Lenovo's system configuration rules for these platforms require processor parity in dual-socket builds — both installed processors must match on model, stepping, and TDP. Mixing the 6226R with a different Gold-tier SKU is not supported.

Pairing this processor kit with matching DDR4 memory, appropriate heatsinks, and a supported ThinkSystem SR590 or SR650 base configuration is required — this is a component option kit, not a standalone server. Consult Lenovo's ServerProven compatibility matrix for validated memory DIMMs and any firmware-level requirements associated with this processor SKU on existing SR590/SR650 installed base.

For server and storage platforms running demanding workloads like network video recorders or AI-driven analytics, processor selection directly impacts channel capacity and inference throughput. The Xeon Gold 6226R sits in a productive tier for mid-density deployments — evaluate the broader Lenovo ThinkSystem processor line if your workload profile demands more cores or a lower TDP ceiling. For infrastructure planning around PoE and switching that feeds these servers, network switch selection is an equally important variable in total system performance.

Frequently Asked Questions

Q: Which Lenovo server platforms is the 4XG7A38082 compatible with?

A: The 4XG7A38082 is qualified for the ThinkSystem SR590 and SR650 server platforms, both of which use the LGA 3647 (Socket P) interface. It is not designed for other ThinkSystem models — verify your chassis model before ordering.

Q: Does the 4XG7A38082 include a heatsink?

A: No. This is a tray-packaged processor option kit. Server platform heatsinks for the SR590 and SR650 are ordered separately. This is standard for server processor option kits intended for system integration.

Q: What is the TDP of the Xeon Gold 6226R in this kit?

A: The Thermal Design Power is 150W. This should be accounted for in chassis cooling, PSU sizing, and rack power planning — particularly in dual-socket SR650 builds where two of these processors may be installed.

Q: Can I install two 4XG7A38082 processors in a dual-socket SR650?

A: Yes. The SR650 supports dual LGA 3647 sockets, and installing matched 6226R processors in both sockets is a supported configuration. Both processors must be identical in model and stepping for proper dual-socket operation.

Q: What is the maximum memory this processor supports?

A: The Intel Xeon Gold 6226R supports up to 1 TB of memory per socket. In a dual-socket SR650, the total addressable memory ceiling reaches 2 TB across both processors.

Q: Is this a new or refurbished processor?

A: The 4XG7A38082 is factory-new, supplied as a genuine Lenovo option kit through commercial distribution channels — not refurbished, not grey-market.

Marty Allison

The 4XG7A38082 is the part number you pull when a ThinkSystem SR590 or SR650 deployment needs a second socket populated or a like-for-like processor replacement at the Xeon Gold 6226R tier. The 150W TDP is the spec I pay most attention to here — it's high enough that thermal planning in the SR650 chassis needs to be deliberate, especially in high-density rack environments where ambient temperatures run warm.

Technical Highlights:

• 16C/32T at 2.9 GHz base: Enough thread count to handle concurrent workloads — virtualized VMS back-ends, analytics inference engines, or multi-tenant application servers — without pushing into the Platinum tier's power envelope.
• 3.9 GHz Turbo Boost: The 1 GHz headroom above base matters for latency-sensitive tasks. Event-triggered VMS processes and licensing services that spike briefly will benefit from boost; sustained bulk encode stays near base.
• Dual UPI links: In a dual-socket SR650, this is what keeps NUMA penalties manageable. Single-UPI dual-socket builds create real bandwidth asymmetry across sockets — the 6226R's dual UPI is the correct choice for any workload that touches memory across both sockets regularly.

Deployment Considerations:

• This ships as a tray part with no heatsink included — the SR590/SR650 chassis heatsink is a separate line item. Don't assume it's bundled; it isn't, and the system won't POST correctly without one properly seated.
• Dual-socket builds require processor parity — mixing the 6226R with a different Gold SKU (even one with matching core count but different TDP or stepping) is an unsupported configuration on these platforms and can trigger firmware warnings or instability.

Best fit: a ThinkSystem SR650 being deployed as a dual-socket analytics or virtualization host in a mid-density rack where the 300W combined processor TDP is manageable and the 32-core / 64-thread aggregate is the right profile for the workload mix — not overkill, not undersized.