Dell PowerEdge R930 4-Bay 2.5"
Configure Your System:
1 Processor Required
Why all four sockets: the E7 platform hangs memory risers, cache, and interconnect off each processor, and the R930's entire purpose is scale-up, one large system instead of several small ones. We configure every unit fully populated, and the choice becomes the shape of the compute.
Higher clocks, fewer cores: pick the lower-core, higher-frequency end of the range when the workload is licensed per core or bound by single-thread speed. Large database engines are the classic R930 tenant, and their licensing often dwarfs the hardware price; every core you do not license is money kept.
More cores, moderate clocks: the high-core-count end of the E7-8800 v4 range puts nearly a hundred cores in one chassis. That is consolidation density nothing else in this generation touches, and for massive parallel workloads and VM farms it is the whole argument.
The E7 line also carries reliability features the E5 platforms do not, deeper error correction and recovery machinery built for systems that are not allowed to go down. If you are pricing an R930, that is usually why.
One honest licensing note before you commit: Windows Server licenses every core in the chassis, and four E7 sockets multiply that math dramatically. Read the OS step before finalizing, and we will run the numbers with you at quote time.
2 Heat Sink Optional
What matters is that this is a 4U chassis cooling four of the largest processors of its generation, plus a wall of memory risers, and it takes airflow discipline seriously. Keep every drive bay blank, riser filler, and internal shroud installed; the R930 cools front to back, and an open path lets air skip the components that need it.
Set expectations accordingly: this is a datacenter machine, engineered for cold-aisle intake air and a rack with real power behind it. It will run in a well-ventilated server room, but it is not a closet server and was never meant to be one.
3 Memory (RAM) 96 DIMM slots, modules added in sets of 4 Required
The guidance that matters at this scale:
Balance across all four processors, always. Each CPU owns its risers, and an unevenly loaded socket starves its share of the workload. Populate in equal sets per processor and fill risers evenly.
Buy it as one matched set, now. This advice matters more on the R930 than on any other server we sell: riser-based memory is sensitive to matching, an upgrade a year later means finding DIMMs that agree with what is installed across dozens of slots, and a mission-critical box does not get casual maintenance windows. Size for the workload's future, not its present.
Capacity is cheap insurance here. The workloads this server carries, large databases, in-memory analytics, massive consolidation, degrade badly when they spill out of RAM, and the whole point of scale-up is not spilling.
Common landing spots: 512 GB (32x 16 GB) as the sensible floor for a machine like this, 1 TB (32x 32 GB) is our default for serious database duty, and beyond that the R930 keeps going where the workload demands it; tell us the target and we will build to it.
4 RAID Controllers Optional
For most R930 deployments the answer is a SAN or external array, and the local controller's job is small and important: a RAID 1 boot pair. The PERC H330 handles that duty honestly and cheaply, and paired with SAN storage it is a complete, correct answer.
When the storage is local, the PERC H730P (2 GB battery-backed cache) is our pick, full stop. A four-socket flagship runs workloads whose write traffic deserves the cache, and against the price of the rest of this build the H730P is a rounding error. The H730 (1 GB) exists between the two and saves little enough that we rarely spec it here.
Running drives passed straight through for a software-defined stack? The list here is RAID controllers only, so call us at 1-800-778-1545 and we will spec the right host bus adapter.
Array guidance for local storage: RAID 10 for database volumes where write latency matters, RAID 6 on anything large, and a hot spare on any array whose rebuild you would rather not race. A server bought for uptime should not have its storage be the casual part.
5 Storage Drives Select up to 4 drives (0/4 Slots Used) Optional
This chassis is built for the classic flagship deployment: the real storage lives on a SAN or external array, and the four local bays exist for a boot volume and little else. A RAID 1 pair of SSDs for the operating system is the standard build, the H330 controller handles it honestly, and the remaining two bays sit empty or hold a second mirrored pair for logs or scratch. Two drives and this page is done.
If you intend to run meaningful local storage, stop and look at the 24-Bay version of this chassis instead; four bays is not a storage subsystem for a machine of this scale, and pretending otherwise wastes the server.
Buy new, even for boot drives. We steer you away from refurbished on drives specifically, and we say that as a company that sells refurbished servers all day. A boot mirror is small money against this build, and a used drive with unknown hours is the wrong place to economize on a mission-critical machine.
One discipline note: keep the boot mirror a boot mirror. The tidiest R930 failures we see are the ones where the OS pair was never asked to do anything else.
6 Remote Access Required
On this server, take Enterprise, and we will be direct about why: an R930 exists to carry workloads a business feels immediately when they stop. The dividing line between the licenses is the remote console, the ability to reach the server's screen, mount an ISO, and rebuild below the OS from anywhere, and on a mission-critical box the scenario that license exists for is not hypothetical, it is the 2 AM page. The Enterprise license costs a fraction of one emergency site visit, and a machine at this tier will see one eventually.
Enterprise also brings automated firmware updates, which matter on a platform with this much firmware surface, and secure erase for the day this server retires holding business-critical data.
One honest note on age: iDRAC8's interface is a product of its era, and some functions lean on older browser technology than iDRAC9 does. It works, we use it daily, but budget a few minutes of setup patience.
If you are deploying 5 or more units, take Enterprise on all of them. Fleet management without remote console access does not work in practice.
7 Power Supply 2x included in build Required
What the arrangement buys you: with four supplies installed, the server runs 2+2 redundancy, meaning the workload survives supply failures and, cabled correctly, the loss of an entire power feed. Run pairs to separate circuits or PDUs and a single upstream failure does not touch the workload; that grid-redundant wiring is the point of the design, and a machine at this tier should be cabled for it.
Every supply is hot-swappable, so a failed unit is replaced without a shutdown, which on this server is the only acceptable way to replace one.
The estimated TDP counter at the bottom of the page will show you where your build lands; on this platform it mostly confirms why the power architecture is standard equipment rather than an option.
8 Network Cards Required
On this server the speed decision is short: a four-socket flagship concentrates an enormous amount of workload onto one network drop, and gigabit networking on an R930 is a mismatch on its face. Treat 10 Gb as the floor: the 2x 10 Gb + 2x 1 Gb combo cards are the standard pick, production on the fast pair, management on the slow pair, and many R930 deployments add PCIe cards beyond that to separate storage, VM, and backup traffic onto their own links.
Then port type, and this is the one that catches people. SFP+ ports take fiber or DAC cables and talk to datacenter-grade switches. BASE-T ports are 10 Gb copper on ordinary RJ45. Neither is better; they must simply match your switch. Order SFP+ against a copper-only switch and the server racks up with nothing to plug into.
An R930's network is usually part of a larger architecture conversation. Call 1-800-778-1545 and we will spec the whole path, switch ports included.
9 Operating System Required
First, an honest compatibility note for this platform. Windows Server 2025 is not certified for this processor generation, and current VMware ESXi releases no longer support it. Plan around Windows Server 2022 and current open-source platforms.
Windows Server 2022 on a four-socket E7 deserves clear eyes: licensing is per core across every core in the chassis, and this chassis can hold nearly a hundred of them. Standard edition's limited virtualization rights make Datacenter edition, with unlimited virtualization rights, the honest comparison for the consolidation duty an R930 is bought for. This licensing math can exceed the hardware price; we will run it with you at quote time before anything is final. On CALs, count your users and round up.
Fair warning on Evaluation Editions: free and fully functional for 180 days, then they expire. Right for labs and proof-of-concept, wrong for production.
VMware: we will install ESXi on request for hosts joining an existing licensed cluster, but we do not recommend starting a new VMware deployment on this platform given the support status.
Proxmox VE deserves a serious look on this chassis: open-source KVM virtualization with clustering and live migration, zero per-core licensing on a server with this many cores, and full current support on this hardware. The economics of a hundred unlicensed cores are hard to argue with.
Ubuntu Server LTS covers Linux database duty with five years of free updates.
No OS is there for teams imaging from their own deployment infrastructure.
A server at this tier deserves a conversation. Tell us the workload at quote time, or call 1-800-778-1545.
WServer Warranty
Dell PowerEdge R930 4-Bay 2.5" Drives [13th Gen] Detailed Review
The Dell PowerEdge R930 4-Bay 2.5" is the refurbished, minimal-local-storage configuration of Dell's 13th-generation 4U, 4-socket flagship. It carries the same platform as the rest of the R930 line: up to four Intel Xeon E7-8800 v4 or E7-4800 v4 processors, 96 DDR4 DIMM slots with a 12 TB memory ceiling, dual PERC support, and up to 10 PCIe Gen3 slots, paired with just four 2.5" hot-swap front bays. It is built for deployments where bulk storage lives on an external SAN or NAS and the local bays only need to cover OS boot, application binaries, and modest local data.
This is the cost-floor R930 when 4-socket compute and the 12 TB memory ceiling are the design drivers and local storage is genuinely minimal. SAN-attached Oracle RAC nodes, SAP HANA appliances with an external storage tier, ERP application servers with the database on a separate platform, and HPC compute nodes backed by an external parallel filesystem are the typical R930 4-Bay deployments. The freed front-bay space is matched by a deep PCIe budget for storage HBAs, Fibre Channel, and high-speed networking.
Every R930 we ship is fully refurbished and tested with a 12+ hour burn-in covering all 96 memory channels, every PCIe slot, and every drive bay, and is backed by a 180-day warranty with extended terms available. To configure a build or price a multi-unit order, call 1-800-778-1545; volume pricing applies at 5 units and above.
When 4 Bays Is the Right Choice
The 4-Bay shares the R930 platform underneath; what differs is the storage strategy. Choose this chassis when:
- Bulk storage is external. Fibre Channel or iSCSI SAN is the primary storage, and the R930 is the compute node. Common for Oracle RAC nodes on shared SAN, SAP HANA appliances with external storage, and ERP application servers with the database elsewhere.
- 4-socket scale and memory are the design drivers, not local disk. The R930's value is its four sockets and 12 TB ceiling. When the workload reads and writes to external storage, four local bays are enough.
- NAS-attached storage covers bulk data. Enterprise NFS or SMB handles capacity; the R930 is the application or database server in front of it.
- External JBOD provides bulk capacity. A PERC H830 plus external SAS shelves delivers large storage outside the chassis, leaving the front bays for boot and logs.
- Acquisition cost matters on the platform. The 4-Bay saves meaningfully versus the 24-Bay while preserving the 4-socket and 12 TB advantage.
If the workload needs substantial local SSD, hybrid SAS plus NVMe tiers, vSAN-class hyperconverged storage, or local IOPS that scale with the application, the 24-Bay variant is the better fit. That tradeoff is covered in Where to Look Instead.
Storage - 4 2.5" Bays
Four 2.5" SAS/SATA hot-swap front bays. This chassis is sized for boot and modest local data, not bulk capacity; bulk storage is expected to live on SAN, NAS, or external JBOD. Unlike the 24-Bay chassis, the 4-Bay does not have dedicated front-bay NVMe infrastructure, so NVMe is added through PCIe add-in cards rather than the front bays.
Common 4-Bay storage profiles
- 2 x SAS SSD boot mirror + 2 x SAS SSD application/log mirror: The volume configuration. Two independent RAID 1 pairs: OS boot, and application data or transaction logs.
- 4 x SAS SSD in RAID 10: Two mirrored pairs striped for a single modest combined boot-and-data array, about 50% capacity efficiency.
- 2 x SAS SSD boot + 2 x NVMe (PCIe add-in): Boot on SAS RAID 1, with a small NVMe hot tier delivered via PCIe cards.
- USB / IDSDM hypervisor boot + 4 x SAS SSD data: ESXi boots from the internal dual-SD module, freeing all four bays for data.
Boot
The R930 has no BOSS module. On the 4-Bay, the volume boot path is a 2-drive front-bay RAID 1 mirror; for hypervisor-only ESXi, internal USB / IDSDM dual-SD mirroring preserves all four bays for data. Because four bays fill quickly, plan the boot strategy at procurement: a boot mirror plus a data mirror consumes all four bays and leaves no room for a hot spare.
Storage Controllers
- PERC H730P (2 GB NV cache, battery-backed): The volume internal controller, supporting RAID 0/1/5/6/10/50/60. See the PERC H730P 2GB cache RAID controller for the part we quote by default.
- PERC H330 (no cache): Entry-tier hardware RAID for a simple boot-and-log mirror set.
- PERC H830 (external): The controller that matters most on a 4-Bay build, fronting external SAS JBOD shelves for the bulk storage that does not live in the chassis.
- 12 Gbps SAS HBA pass-through: For software-defined storage where the OS or hypervisor owns the disks.
The platform tops out at the H730P generation; the 8 GB-cache H740P is a 14th-gen part and is not available here.
Processors
The 4-Bay takes the same 2 or 4 Intel Xeon E7-8800 v4 or E7-4800 v4 (Broadwell-EX) processors as the rest of the R930 line. The E7-8800 family supports 8-socket-capable QPI interconnect (the R930 uses four sockets); the E7-4800 family is 4-socket-only at a lower price. At matching tiers the two lines share identical core counts, clocks, and TDPs, so for almost every build the E7-4800 v4 SKU is the right call; E7-8800 v4 only makes sense when a forward path to an 8-socket platform is planned, which is rare.
Common SKU choices
- E7-8890 v4 / E7-4890 v4 (24 cores, 2.2 GHz, 165W): Maximum core count. 4-socket = 96 cores / 192 threads.
- E7-8880 v4 / E7-4880 v4 (22 cores, 2.2 GHz, 150W): High-core balanced choice. 4-socket = 88 cores.
- E7-8870 v4 / E7-4870 v4 (20 cores, 2.1 GHz, 140W): Higher-core balanced, strong for SQL Server consolidation.
- E7-8860 v4 / E7-4860 v4 (18 cores, 2.2 GHz, 140W): Mid-range volume SKU. 4-socket = 72 cores.
- E7-8855 v4 / E7-4850 v4 (16 cores, 2.1 GHz, 115W): Lower-tier balanced. 4-socket = 64 cores.
- E7-8830 v4 / E7-4830 v4 (14 cores, 2.0 GHz, 115W): Cost-efficient mid-range.
- E7-8820 v4 / E7-4820 v4 (10 cores, 2.0 GHz, 115W): Entry-tier 4-socket.
Top-bin E7 parts at 150-165W generate substantial heat across four sockets; confirm fan and ambient planning at quote time.
Memory
Eight memory risers, each with 12 DDR4 DIMM slots: 96 slots and 32 channels across four sockets, running at 2400 MT/s with E7 v4 processors, identical to the 24-Bay. Maximum capacity is 12 TB using 128 GB LRDIMMs across all 96 slots; 128 GB modules are LRDIMM-only, and most builds use 64 GB LRDIMMs (6 TB maximum) for better cost per GB. Memory mirroring and memory failover are supported and trade capacity for RAS.
Practical memory configurations
- 1 TB (16 x 64 GB LRDIMM): Dense virtualization or SQL Server consolidation where 1 TB covers the workload.
- 2 TB (32 x 64 GB LRDIMM): Larger in-memory database or ERP host; SAP HANA scale-up small/medium tier.
- 3 TB (48 x 64 GB LRDIMM): SAP HANA mid-tier appliance or large Oracle RAC node.
- 4 TB (64 x 64 GB LRDIMM): Large in-memory analytics platform.
- 6 TB (96 x 64 GB LRDIMM, fully populated): Maximum capacity with 64 GB modules.
- 12 TB (96 x 128 GB LRDIMM, fully populated): The platform ceiling, for large SAP HANA appliances and in-memory scale-up.
Intel Optane Persistent Memory is not supported on the R930; the E7 v4 platform predates Cascade Lake's PMem support.
Networking and PCIe Expansion
Networking starts with a Network Daughter Card (NDC) in its own slot, so it does not consume a general-purpose PCIe slot. Beyond the NDC, the R930 offers up to 10 PCIe Gen3 slots plus a dedicated RAID slot. On a 4-Bay build this PCIe budget is the heart of the configuration: with bulk storage external, the slots carry the storage and networking fabrics.
- Storage HBA capacity: A PERC H830 plus external JBOD shelves for direct-attached scale-out. Typically 2 to 4 slots with redundancy.
- Fibre Channel: Dual 16 Gbps or 32 Gbps FC HBAs for SAN connectivity, usually 2 slots.
- High-speed networking: Dual 25 GbE or quad 10 GbE NICs for production and storage networks, 1 to 2 slots.
- InfiniBand: ConnectX-class HCAs for HPC fabrics, 1 to 2 slots.
- NVMe add-in cards: PCIe NVMe SSDs for a local hot tier when the four front bays are not enough, 1 to 4 slots.
Exact slot counts and widths depend on riser selection and CPU population; we confirm the PCIe layout for your configuration at quote time.
GPU Support
The R930 is a scale-up compute and memory platform, not a GPU server, and that is even more true of the storage-light 4-Bay, whose PCIe slots are usually committed to storage and networking fabrics. The 4U chassis can physically house accelerator cards within the thermal and power envelope of a fully populated 4-socket system, but Dell did not position the R930 as a GPU platform. For GPU-bound AI/ML training or large-scale inference, a purpose-built accelerator platform such as a 14th-generation R740xa or newer is the right tool. For a specific single-card or FPGA case, tell us at quote time and we will confirm support.
Management - iDRAC8 Generation
iDRAC8 Enterprise with Lifecycle Controller: remote KVM, virtual media, power management, hardware monitoring, and OpenManage integration, identical to the rest of the R930 line and the same platform as the R630, R730, and R830. Relative to iDRAC9 on 14th-gen platforms, iDRAC8 lacks Silicon Root of Trust and System Lockdown. For the full iDRAC8 walkthrough, see our Dell PowerEdge R630 10-Bay 2.5" page.
Power and Cooling
Four hot-swap Platinum power supplies in 750W or 1100W, configured for 2+2 (or 3+1) redundancy. Power draw on a 4-Bay build runs lower than the 24-Bay because far fewer drives are active, typically 200 to 400W less depending on drive count and type.
| Workload profile | Typical draw | PSU recommendation |
|---|---|---|
| Light: 2-socket, 512 GB RAM, 4 SSDs, 10 GbE + FC HBA | 350-520W | 4 x 750W Platinum (2+2) |
| Balanced: 4-socket, 2 TB RAM, 4 SSDs, 25 GbE + dual FC | 800-1200W | 4 x 750W or 4 x 1100W Platinum (2+2) |
| Heavy: 4-socket E7-4860 v4, 4 TB RAM, 4 SSDs, dual 25 GbE + storage HBA + JBOD | 1200-1800W | 4 x 1100W Platinum (2+2) |
| Maximum: 4-socket E7-4890 v4, 12 TB RAM, 4 SSDs, 25 GbE + InfiniBand + storage | 1900-2500W | 4 x 1100W Platinum (2+2) |
1100W PSUs are the volume specification for 4-socket builds regardless of chassis variant; 750W suits 2-socket or lighter 4-socket loads. Confirm rack PDU and cooling capacity, and remember that four high-TDP CPUs and up to 12 TB of memory put real heat into 4U even with only four drives.
Physical Specs and Platform Notes
- Form factor: 4U rack chassis, four-socket scale-up design. Plan full-depth rail and cable-management-arm clearance in the rack.
- PCIe expansion: Up to 10 PCIe Gen3 slots plus a dedicated RAID slot and a separate NDC slot; full-height cards supported, with the exact layout set by riser and CPU population. On the 4-Bay this budget typically carries storage HBAs and SAN or network fabrics.
- Parts availability: Mature but thinner than the volume 2-socket platforms. The R930's smaller installed base means E7-8800/E7-4800 v4 CPUs and R930-specific FRUs are less abundant on the secondary market than R630/R730 parts. Dell ProSupport for the platform is at end-of-service; third-party maintenance is the standard production support path in 2026.
- Accessories we recommend: The PERC H730P 2GB cache RAID controller for the internal bays, a PERC H830 for external JBOD when bulk storage is direct-attached, plus a 4U-depth ReadyRails kit. The 1U and 2U rail kits and bezels elsewhere in our catalog do not fit the 4U R930 chassis; we match the correct 4U hardware at quote time.
- Platform notes: No BOSS module, so boot is via front-bay RAID 1 or USB/IDSDM; no dedicated front-bay NVMe (NVMe via PCIe add-in only); iDRAC8 without Silicon Root of Trust; a PCIe Gen3 ceiling; no Optane PMem; verify rack PDU capacity for four 1100W PSUs.
Our Assessment
Where it excels: The R930 4-Bay is the right call when the platform's 4-socket compute and 12 TB memory capacity matter and local storage is genuinely minimal because bulk data lives outside the chassis. SAN-attached Oracle RAC nodes, SAP HANA appliances with an external storage tier, ERP application servers with an external database, HPC compute nodes on an external parallel filesystem (Lustre, GPFS, BeeGFS), virtualization hosts backed by external shared storage, and dense application-server consolidation in front of a SAN are where it shines.
Where to look instead: If the workload needs substantial local IOPS or a hybrid SAS-plus-NVMe tier, the R930 24-Bay is the variant to choose. If 2-socket E5-2600 v4 covers the compute, the R730 and 1U R630 cost far less. If four sockets are needed but 3 TB and 2U suffice, the R830 8-Bay saves rack space and money. Workloads needing more than 12 TB, Optane PMem, or iDRAC9 firmware integrity are a 14th-gen R940 conversation.
Bottom line: The 4-Bay versus 24-Bay decision comes down to storage strategy: external-first (4-Bay) or hybrid local-plus-external (24-Bay). For a SAN- or NAS-backed 4-socket compute appliance in Dell's 13th generation, the R930 4-Bay is the cost-correct answer, and it remains a sound buy in 2026 where the 12 TB ceiling and E7 v4 RAS fit the requirement. We help match CPU, memory tier, HBA, and PCIe slot allocation to your existing storage architecture at quote time.
Honest Limitations
- Four SFF bays is the chassis ceiling. It cannot be field-converted to 24-Bay; the backplane and drive cage are 4-Bay-specific. Growth in local storage requires external attachment via a PCIe HBA and JBOD shelves.
- No dedicated front-bay NVMe. Unlike the 24-Bay, this chassis lacks NVMe-capable front bays; NVMe is possible only through PCIe add-in cards.
- Four bays fill immediately. A boot mirror plus an application or log mirror uses all four bays, leaving no room for a hot spare without giving up a mirror.
- Storage must be externalized. The 4-Bay assumes SAN, NAS, or external direct-attached storage, and the PCIe budget is allocated accordingly.
- All R930 platform limits apply. 12 TB memory ceiling, DDR4 2400 MT/s, no Optane PMem, no 14th-gen successor stocked here, 4U rack space, and four 1100W PSUs at full 4-socket load. See our R930 24-Bay page for the full discussion.
- 13th-gen platform constraints apply. iDRAC8 without Silicon Root of Trust, no BOSS module, PERC H730P as the top controller with no H740P, a PCIe Gen3 ceiling, and Dell ProSupport at end-of-service. See our R630 10-Bay page for the full 13th-gen discussion.
- Thinner parts availability. The R930's smaller installed base means E7 v4 CPUs and chassis-specific FRUs are less abundant on the secondary market than R630/R730 parts.
- OS support narrowing. Modern OS releases may have limited 13th-gen support; verify compatibility for deployment horizons beyond 2026.
Workload Fit
| Excels at | Where to look elsewhere |
|---|---|
| SAN-attached Oracle RAC nodes | Substantial local storage IOPS needed (use R930 24-Bay) |
| SAP HANA with an external storage tier | Local NVMe or hybrid local storage (use R930 24-Bay) |
| ERP servers with an external database | 2-socket sufficient (use R630/R730 at much lower cost) |
| HPC nodes with an external parallel filesystem | 4-socket-in-2U sufficient (use R830) |
| Virtualization hosts with external shared storage | More than 12 TB memory (use R940) |
| Dense app-server consolidation backed by SAN | iDRAC9 firmware integrity required (use R940) |
| Cost-floor R930 compute appliance | Local-storage-heavy workloads (use R930 24-Bay) |
Where to Look Instead
- Same platform, dense local storage: the Dell PowerEdge R930 24-Bay 2.5" adds 20 more SFF bays and up to 8 front-bay NVMe for workloads where local storage scales with the application.
- Four sockets in 2U: the Dell PowerEdge R830 8-Bay 2.5" is the same-generation 4-socket platform with E5-4600 v4, 48 DIMM slots, and a 3 TB ceiling. It saves rack space and cost when 3 TB is enough.
- Two-socket, lower cost: the Dell PowerEdge R730 8-Bay 3.5" and the 1U Dell PowerEdge R630 10-Bay 2.5" cover workloads that fit within 2-socket E5-2600 v4 and up to 1.5 TB, at a much lower price.
- 13th-gen LFF capacity: the Dell PowerEdge R730xd 24-Bay 2.5" and its 12-Bay 3.5" companion are the dense-storage alternative when capacity drives, not socket count, are the requirement.
- Default storage controller: the PERC H730P 2GB cache RAID controller is the controller we quote for most R930 builds.
- Next generation: the 14th-gen R940 is the forward path for iDRAC9, faster DDR4, Optane PMem, and a longer support runway. We do not stock it but can source it on request. There is no direct HPE 4-socket 4U counterpart in our current catalog; the closest concept is the ProLiant DL580 Gen9, which we can also source on request.
Ready to Configure?
Tell us your workload (SAN-attached Oracle RAC, SAP HANA with external storage, ERP application server, HPC compute node, virtualization host with shared storage), target socket count (2 or 4), CPU preference, memory capacity in the 1-12 TB range, local storage configuration (two RAID 1 mirrors, 4-drive RAID 10, or an alternative), external storage architecture (Fibre Channel, iSCSI, NAS, or direct-attached JBOD), PCIe expansion needs (storage HBA, FC HBA, networking, InfiniBand), and quantity. We respond within 24 hours.
For SAN-attached deployments, share your existing storage architecture, including vendor, fabric speed, and expected per-node bandwidth, and we will configure FC or iSCSI HBA selection and PCIe slot allocation accordingly.
Every Wholesale Servers R930 ships after a 12+ hour burn-in covering every PCIe slot, all 96 memory channels, and every drive bay. The standard 180-day warranty is included, with 1-Year, 2-Year, and 3-Year Premium options available. Call 1-800-778-1545 or use the quote form on this page; volume pricing applies at 5 units and above.
Save Your Design
Click the Add to Quote button at the bottom of your screen to save your design as a draft order for future reference and to check for discounts, lead time, and availability. Most servers ship within 1-3 days.
