Deploy NVMe-over-Fabrics without enterprise cost barriers. RocketStor 4243AS delivers 100GbE performance, low latency, and flexible NVMe/TCP or RoCE storage. Overview Specifications Downloads Resources Enclosure Specifications Form Factor 2U Rack Mount Front Drive Bays 24 x 2.5-Inch NVMe SSD Trays Fabric Adapter Slots 2x QSFP28 (100Gbps) Fabric Adapter Western Digital RapidFlex C2000 Fabric Bridge Adapter Cabling Passive (1 - 5m) and Active Optical (5m - 20m) Power Supply 2 x 1300w 100-240VAC Platinum, CRPS, Hot-Plug capable PSUs Platform Management Controller ARM Based BMC (Aspeed 2620) Remote Management 1Gb Ethernet Connector (RJ45), BMC remote management Environmental Working Temperature: 10°C ~ +35°C Storage Temperature: 5°C ~ +45°C Operating Voltage: 100V ~ 240V AC In; 1300W RocketStor 4243AS: The Entry-Point to NVMe-over-Fabrics Eliminating Cost Barriers for the Software-Defined Enterprise The RocketStor 4243AS is HighPoint’s premier Entry-Level NVMe-oF Storage Solution, designed to bring the power of Composable Disaggregated Infrastructure (CDI) to the wider industry. By removing the "Enterprise Tax" and high-budget hurdles associated with complex, over-engineered multi-controller arrays, the RS4243AS makes high-density, Fabric-Attached Block Storage accessible to mid-market enterprises, regional AI pods, and media production houses. Disaggregated Server Storage - Redefined Hardware-Accelerated Fabric Offload (WDC RapidFlex™ C2000) At the heart of the RocketStor 4243AS is the Western Digital RapidFlex™ C2000, a dedicated NVMe-oF Fabric Bridge that transforms the storage enclosure into a high-performance network target. Near-Zero Target Overhead: Unlike software-defined targets that consume massive CPU cycles to manage network traffic, the C2000 offloads the entire NVMe-oF protocol stack to dedicated silicon. This ensures the RS4243AS delivers consistent 100GbE line-rate performance without taxing the enclosure's management resources. 120-Host Connectivity: Consolidate your storage and serve up to 120 independent host initiators simultaneously. The C2000 manages these virtual controllers in hardware, providing seamless multi-tenant access without protocol bottlenecks. Transparent Hardware Bridge: The RS4243AS functions as a pure hardware bridge, presenting remote NVMe storage to the network with the same simplicity and speed as locally attached drives. Unmatched Fabric Flexibility: RoCE v2 or TCP The RocketStor 4243AS is designed to adapt to your existing infrastructure while providing a clear path for future performance upgrades. NVMe/TCP (Standard Ethernet): Deploy instantly over your current standard Ethernet network. Experience the benefits of disaggregated storage with zero specialized hardware and maximum compatibility across your existing switch fabric. NVMe/RoCE v2 (RDMA): For workloads requiring sub-microsecond latency—such as AI/ML training and 8K video editing—leverage RoCE v2 to achieve "wire-speed" performance. Note: Maximum RDMA performance requires a RoCE-capable SmartNIC on the host initiator. Green Power & Thermal Reliability: Optimized for Efficiency The RS4243AS is engineered for the "Green" data center, where every watt of power and every degree of heat matters. "Lane-per-Drive" Architecture: Utilizing the HighPoint R1528D PCIe Switch , the RS4243AS allocates a dedicated Gen4 x1 lane to each of the 24 NVMe bays. This provides ~45GB/s of internal bandwidth—perfectly saturating the Dual 100GbE network interface (~25GB/s)—without generating the massive heat of an over-provisioned 96-lane configuration. Reduced Thermal Load: Running 24 lanes instead of 96 significantly lowers the internal temperature of the chassis, protecting your NVMe media from thermal throttling and extending hardware lifespan (MTBF). Operational Savings: A cooler-running system requires less active cooling, directly reducing your monthly electricity and facility cooling costs. The "Zero-Constraint" Hardware Bridge The RS4243AS offers a level of simplicity and open-standard compatibility that software-heavy arrays cannot match. No Drivers, No Constraints: Natively supported by any OS with an NVMe-oF initiator (Linux, Windows Server 2025, VMware). RedFish Open Management: Utilize the integrated ARM-based BMC and native Redfish® API for seamless, out-of-band (OOB) integration into your existing orchestration and monitoring dashboards. True BYOD (Bring Your Own Drive) Freedom Escape vendor lock-in and high-margin proprietary drives. The RS4243AS is a True Open Platform: Vendor Agnostic: Mix and match U.2/U.3 SSDs from Samsung, Micron, Sandisk, or Solidigm based on your specific workload and budget requirements. Target Industry Solutions Collaborative 4K/8K Video Scratch Space for high-speed post-production. High-efficiency block storage for Post-VMware / Proxmox migrations and SDS nodes. High-speed data ingestion and "Fast Buffer" storage for regional AI pods with limited power/cooling envelopes. Explore the RocketStor 4243AS Specialized NVMe-oF Hardware for RoCE & TCP Workflows (1) 2U Rack Mount Chassis Industry standard form factor Designed for Industrial/Enterprise deployment (2) 24 Independent Drive bays Industry standard 2.5" form factor Designed for U.2 NVMe storage media (1) Integrated Storage Adapter HighPoint Rocket 1528D Switch Adapter Allocates x1 dedicated lanes to each 2.5" drive bay Intelligent Firmare Layer Monitors hardware telemtery in real-time: power consumption drive status host connection status (3) Remote Management port OOB, BMC, Redfish (2) Integrated NVMe over Fabric Controller WDC RapidFlex C2000 Fabric Bridge Adapter 2x 100 GbE ports - Dual 100GbE Line-Rate Connectivity Hardware-Accelerated NVMe-oF Offload RoCE & TCP Workflows Driving Innovation Through Strategic Engineering Partnerships The RocketStor 4243AS NVMe-oF Hardware Storage Platform has been validated by a rage of industry leading manufactures, developers and solution providers. Forging Strategic Partnerships: Your Software. Our Hardware Platform. HighPoint is actively seeking collaboration with Software-Defined Storage (SDS) and CDI Orchestration providers. Leverage our hardware-accelerated fabric offload to offer your customers disruptive economics and validated performance. Chat with an Expert News Shifting the Storage Paradigm: Why Out-of-Band BMC Architecture is Non-Negotiable for Next-Gen NVMe-oF Pools Unlocking Infrastructure Agility: The Synergy of BMC and Redfish API in Composable Disaggregated Infrastructure Stop Wrestling with "Dumb" JBODs: A Modern Blueprint for Deterministic NVMe-oF Scaling HighPoint and OSNexus Launch Turnkey True Unified Storage Solution to Accelerate AI, M&E, and Private Cloud Infrastructure HighPoint Partners with Insyde® Software to Deliver Secure, OOB OpenBMC Management for Disaggregated NVMe-oF Storage Platform The Partnership Advantage: Scaling Composable Storage with Zero-Driver Engineering Solving the AI Storage Bottleneck: An Architectural Deep-Dive into the RocketStor 4243AS Breaking the Local Storage Myth: Why RoCE and the RocketStor 4243AS are the New Standard for Disaggregated Storage Solving the AI Data Stalling Problem: Why Your Inference Cluster Needs a CDI Storage Tier HighPoint Unveils New Category of CDI Hardware Storage Platforms; Launches RocketStor 4243AS NVMe-oF Solution for Liquid Infrastructure The Death of "Stranded Capacity"—Why Your NVMe Storage is Only 60% Efficient Choosing Your Path NVMe/TCP vs. RoCE v2 for the Modern Fabric HighPoint NVMe Switch Adapters and Systems from our Partners Software & Drivers SSDP Utility Download Product Documents Quick Reference Guide Download Quick Installation Guide Download Compatibility List Download Hardware User Guide Download BMC User Guide Download Installation Instructions NJ21A59-26 Download Lossless Test Direct Attach Global Pause Download RAID Controller/Enclosure Port Guide Download Datasheet

As a leading company in RAID storage solutions, HighPoint Technologies, Inc. develops and manufactures connectivity solutions for professional and industrial clients. 7604 learn more Test Drive 7604 learn more Test Drive 7604 learn more Test Drive 7604 learn more Test Drive 7604 learn more Test Drive NEWS Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Video Title Describe your video here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Video Title Describe your video here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Video Title Describe your video here Image Title Describe your image here Image Title Describe your image here Image Title Describe your image here Video Title Describe your video here HighPoint Unveils Comprehensive PCIe Gen 5 MCIO Expansion Ecosystem to Eliminate Slot Starvation in Dense AI & HPC Infrastructure Switch-Based vs. Retimer-Based MCIO Adapters: Architectural Comparison Unlocking PCIe Expansion in Dense AI Infrastructure Shifting the Storage Paradigm: Why Out-of-Band BMC Architecture is Non-Negotiable for Next-Gen NVMe-oF Pools HighPoint Launches the Rocket 7602L: The Industry’s Premier Bootable Dual-Port PCIe Gen5 M.2 AIC for Workstations and Edge Servers Bypassing the Pre-Boot Wall: Why Native Motherboard M.2 Slots Can't Handle Redundant OS RAID HighPoint Unveils Industry’s Most Comprehensive PCIe Gen5 P2P and GPU Direct Storage (GDS) Solution Portfolio for White-Box AI and Edge Infrastructure How to Decouple PCIe Gen5 Slots from Rigid Motherboard Layouts in Custom Enclosure HighPoint Unveils Comprehensive PCIe Gen 5 MCIO Expansion Ecosystem to Eliminate Slot Starvation in Dense AI & HPC Infrastructure Switch-Based vs. Retimer-Based MCIO Adapters: Architectural Comparison Unlocking PCIe Expansion in Dense AI Infrastructure Shifting the Storage Paradigm: Why Out-of-Band BMC Architecture is Non-Negotiable for Next-Gen NVMe-oF Pools HighPoint Launches the Rocket 7602L: The Industry’s Premier Bootable Dual-Port PCIe Gen5 M.2 AIC for Workstations and Edge Servers Bypassing the Pre-Boot Wall: Why Native Motherboard M.2 Slots Can't Handle Redundant OS RAID HighPoint Unveils Industry’s Most Comprehensive PCIe Gen5 P2P and GPU Direct Storage (GDS) Solution Portfolio for White-Box AI and Edge Infrastructure How to Decouple PCIe Gen5 Slots from Rigid Motherboard Layouts in Custom Enclosure HighPoint Unveils Comprehensive PCIe Gen 5 MCIO Expansion Ecosystem to Eliminate Slot Starvation in Dense AI & HPC Infrastructure Switch-Based vs. Retimer-Based MCIO Adapters: Architectural Comparison Unlocking PCIe Expansion in Dense AI Infrastructure Shifting the Storage Paradigm: Why Out-of-Band BMC Architecture is Non-Negotiable for Next-Gen NVMe-oF Pools HighPoint Launches the Rocket 7602L: The Industry’s Premier Bootable Dual-Port PCIe Gen5 M.2 AIC for Workstations and Edge Servers Bypassing the Pre-Boot Wall: Why Native Motherboard M.2 Slots Can't Handle Redundant OS RAID HighPoint Unveils Industry’s Most Comprehensive PCIe Gen5 P2P and GPU Direct Storage (GDS) Solution Portfolio for White-Box AI and Edge Infrastructure How to Decouple PCIe Gen5 Slots from Rigid Motherboard Layouts in Custom Enclosure HighPoint Unveils Comprehensive PCIe Gen 5 MCIO Expansion Ecosystem to Eliminate Slot Starvation in Dense AI & HPC Infrastructure Switch-Based vs. Retimer-Based MCIO Adapters: Architectural Comparison Unlocking PCIe Expansion in Dense AI Infrastructure Shifting the Storage Paradigm: Why Out-of-Band BMC Architecture is Non-Negotiable for Next-Gen NVMe-oF Pools HighPoint Launches the Rocket 7602L: The Industry’s Premier Bootable Dual-Port PCIe Gen5 M.2 AIC for Workstations and Edge Servers Bypassing the Pre-Boot Wall: Why Native Motherboard M.2 Slots Can't Handle Redundant OS RAID HighPoint Unveils Industry’s Most Comprehensive PCIe Gen5 P2P and GPU Direct Storage (GDS) Solution Portfolio for White-Box AI and Edge Infrastructure How to Decouple PCIe Gen5 Slots from Rigid Motherboard Layouts in Custom Enclosure Product Categories NVMe NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe Switch Adapter NVMe Switch AIC NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) NVMe RAID Adapter NVMe RAID AIC (M.2) NVMe RAID AIC (E1.S) Ext. Storage Enclosure NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) NVMe AIC SSD (Workstation&Server) NVMe AIC SSD (Mac) SAS/SATA RAID SAS/SATA HBA USB Cable 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 12G SAS RAID 6G SAS RAID 6G SATA RAID 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G Storage Enclosure 6G Storage Enclosure 12G SAS HBA 6G SATA HBA USB 20G USB 10G USB 5G 12G SAS RAID 6G SAS RAID 6G SATA RAID 12G Storage Enclosure 6G Storage Enclosure NVMe SAS/SATA USB Communities Stay up to date with the latest information about HighPoint and our updates for the operating systems. Our communities show the supported versions for each product, FAQs and relevant update notes. Linux Community > Windows Community > Mac Community >

Sep 2026 Download HighPoint Launches Gen5 AI Upgrade Solutions: Bringing Blackwell GPU Compute and Gen5 Storage to Edge Infrastructure Technical dedicated Solutions Page Enables x86 Edge Infrastructure to Run 600W Blackwell Accelerators and Gen5 Storage at Full 64 GB/s Line Rate Without Full Platform Refreshes. FREMONT, CA — September 2026 — HighPoint Technologies, Inc., a pioneer in high-performance PCIe switching architectures and NVMe storage connectivity, has unveiled its new portfolio of Gen5 AI Upgrade Solutions . Engineered for systems integrators and AI infrastructure professionals, the new dedicated landing page details deployment architectures for adding NVIDIA Blackwell RTX 6000 Pro GPUs and high-density Gen5 NVMe arrays to existing x86 Intel and AMD systems via autonomous hardware-level P2P switching fabric. Overcoming the Physical and Bus Bottlenecks of Edge AI As large language models, multimodal vision pipelines, and real-time inference expand into decentralized edge environments, engineering teams face critical infrastructure roadblocks: The Gen4 Replacement Dilemma: Swapping out entire server backbones for native Gen5 hardware demands significant capital expenditure and introduces extensive project delays driven by persistent market supply constraints for enterprise Gen5 CPUs and DDR5 memory. The Physical Edge Barrier: Compact 1U/2U server chassis and short-depth edge enclosures often face severe slot limitations, insufficient onboard power, and thermal constraints that physically prevent housing 600W active-cooled GPUs internally. HighPoint’s Gen5 AI Upgrade Solutions resolve both challenges at the silicon layer, unlocking native NVIDIA GPUDirect P2P and GPUDirect Storage (GDS) compatibility across x86 Intel and AMD platforms. Powered by an integrated 48-lane Broadcom PCIe Gen5 switch and PCI-SIG CopprLink connectivity, these solutions establish a localized, autonomous downstream sub-fabric. High-throughput AI workloads execute via hardware-level Peer-to-Peer (P2P) DMA at full line-rate (up to 64 GB/s bi-directional)—transferring data directly between local NVMe storage, edge accelerators, and GPUs while completely bypassing the host CPU, system RAM, and host motherboard bus “Edge AI architects are caught between two bad options: replace entire server fleets prematurely, or sacrifice GPU performance due to host bus bottlenecks,” said May Hwang, Marketing Director at HighPoint Technologies. “Our Gen5 AI Upgrade Solutions offers an immediate third path. By combining dedicated 64 GB/s CopprLink cabling with our autonomous 48-lane switching fabric, system builders can integrate Blackwell-class compute and Gen5 storage into their existing x86 Intel and AMD edge footprints at full line rate.” Three Targeted Architectural Deployment Models The newly launched solution details three modular hardware deployment blueprints tailored to real-world edge environments: 1. Disaggregated External Compute: Couples the Rocket 7634D/7638D/1638E host adapter with the RocketStor 8631D-1300W enclosure over dedicated CopprLink™ CDFP x16 cabling, offloading 1300W of power and active cooling outside 1U/2U servers to run 600W Blackwell GPUs at continuous peak boost clocks. 2. Vertical / Cascading PCIe Expansion: Bridges the Rocket 1600A Series to secondary adapters like the Rocket 1528D via high-density MCIO connections, multiplying a single host x16 slot into an expansive hierarchy supporting 8+ NVMe drives alongside edge GPUs and NPU accelerators without chassis overhauls. 3. Horizontal Direct PCIe Expansion: Connects GPUs and Gen5 NVMe scratch pools directly to a single-tier crossbar matrix (such as the Rocket 7638D ), delivering zero-hop, ultra-low latency (~100–150 ns) for live inference, token caching, and prompt fine-tuning. Learn More To explore architectural diagrams, request evaluation units, and access technical blogs comparing CopprLink™ vs. OCuLink, please visit HighPoint’s Gen5 AI Upgrade Solutions .

Selecting the right PCIe Gen 5 MCIO expansion adapter determines your system’s scaling limits, drive density, host compatibility, and virtualization boundaries. While both switch-based and retimer-based adapters leverage high-density Mini Cool Edge IO (MCIO / SFF-TA-1016) cabling, they solve fundamentally different architectural challenges. Examining the Architectural Differences Switch-Based Adapters (Rocket 1628A, Rocket 1624A) · Unified Compute & Storage Expansion: A dedicated 48-Lane Broadcom PCIe Gen 5 switch IC and embedded processor enables these adapters to function as autonomous PCIe fabrics. Pairing these adapters with MCIO-PCIEX16-G5 bridge cards can transform each MCIO port into physical PCIe 5.0 x16 slot: up to 4x slot at x8 speeds or 2x full-bandwidth x16 slots – ideal for adding GPUs, DPUs, FPGAs, and 400GbE NICs. · Massive Drive Density: The flagship Rocket 1628A’s four MCIO 8i ports directly support up to 8 NVMe SSDs (U.2, U.3, E3.S, E1.S) or up to 32 NVMe drives via UBM/VPP-compliant backplanes. · No Host Bifurcation Required: These adapters are designed to operate in any standard PCIe x16 slot without relying on motherboard BIOS bifurcation settings. · Peer-to-Peer (P2P) Direct Transfer: Provides direct data pathways between all hosted PCIe devices (GPUs, DPUs, NICs, and NVMe drives) bypassing the host CPU and system DRAM to eliminate latency. 1. Retimer-Based Adapters (Rocket 1624L, Rocket 1624M) · Signal Conditioning & Extension: Retimers adapters sample, re-time, and re-equalize PCIe 5.0 signals (32 GT/s) across longer trace lengths and cables, enabling PCIE cards to be mounted far away from the host CPU without compromising the signal quality or introducing latency · Point-to-Point PCIe Slot & Storage Extension: Each adapter is equipped with two MCIO 8i ports, which direct support up to four NVMe drives, or pair with an MCIO-PCIEX16-G5 bridge card to relocate 1–2 remote PCIe slots within the host chassis. · Host Bifurcation Requirement: The host motherboard and BIOS must support PCIe bifurcation (such as x8/x8 or x4/x4/x4/x4) to allocate lanes to each downstream device. PCIe Slot Expansion A major advantage of the MCIO ecosystem is the ability to bypass rigid physical risers and build flexible compute, networking, or hybrid configurations using HighPoint’s MCIO-PCIEX16-G5 expansion bridge card. · Switch-Based Multi-Slot & Hybrid Fabric (Rocket 1628A / 1624A): The 48-lane switch allows system builders to mix and match workloads on a single card. For example, the Rocket 1628A can allocate 2x MCIO ports to drive a full-bandwidth GPU/accelerator slot via a bridge card, while dedicating the remaining 2x MCIO ports to an ultra-dense 16-to-32 drive NVMe backplane or auxiliary 400GbE NICs. · Retimer Point-to-Point Extension (Rocket 1624L / 1624M): Provides direct 1:1 physical lane passthrough. Connecting both MCIO 8i ports to the MCIO-PCIEX16-G5 bridge card creates an extended PCIe x16 slot, which can be relocated in a cooler or more spacious zone within the server chassis (provided the host slot supports x8/x8 bifurcation). Which Solution is best for Your Workload? Choose a Retimer Solution (Rocket 1624L / 1624M) when budget efficiency is a primary concern, the host motherboard BIOS natively supports PCIe bifurcation, and the target build requires clean point-to-point routing for up to 4 NVMe SSDs or 1–2 remote PCIe slots (via MCIO-PCIEX16-G5 bridge cards). The ability to rebuild PCIe signals make these adapters excellent solutions for unique chassis layouts, as they enable NVMe drives and PCIe cards to be located further away from the host CPU without introducing significant latency. Choose a Switch Solution (Rocket 1628A / 1624A) when the workload requires maximized performance and scalability. The Switch architecture can support up to 32 NVMe SSDs via enterprise backplanes, or expand a single host slot into two PCIe 5.0 x16 or four PCIe 5.0 x8 slots via MCIO-PCIEX16-G5 bridge cards - all without requiring host BIOS bifurcation These adapters also provide direct, P2P (peer-to-peer) pathways between hosted, effectively offloading any associated CPU and memory overhead from the host platform; ideal for applications that require GDS (GPU direct Storage) capability. Switch-based adapters remain the definitive choice for enterprise environments, where advanced storage and connectivity features - such as NVMe hot-plug/hot-swap support, UBM/VPP sideband management, and real-time device health monitoring - are essential. Switch adapters are designed to ensure maximum fabric flexibility and universal compatibility across servers with locked or limited bifurcation settings. Comprehensive Feature & Specification Comparison Architectural Feature Retimer-Based (Rocket 1624L / 1624M) Switch-Based (Rocket 1628A/1624A) Host Interface PCIe 5.0 x16 PCIe 5.0 x16 PCIe Layer Operation Layer 1 (Physical Retiming) Layers 1, 2, 3 (Packet Routing) Total MCIO 8i Ports 2 R1624A: 2 R162A8: 4 PCIe Slot Expansion Via MCIO-PCIEX16-G5 (1x x16 or 2x x8) Via MCIO-PCIEX16-G5: R1624A: 1x x16 or 2x x8 R1628A: 2x x16 or 4x x8) Max Direct NVMe Drives Up to 4 Drives (U.2/U.3/E3.S) R1624A:4 R1628A: 8 (U.2/U.3/E3.S) Max Backplane NVMe Support Up to 4 Drives Up to 16 (R1624A) / Up to 32 (R1628A) (UBM Backplane) Motherboard Bifurcation Mandatory (Host must split lanes) Not Required (Hardware Switched) IOMMU & ACS Control Supported Supported Direct P2P Accelerator Offload Host CPU Mediated Yes (Local Switch Fabric) Primary Deployment Focus Cost-effective point-to-point cable/slot extension Switched AI compute & mid-tier storage Learn More

Modern AI, machine learning, and high-performance computing (HPC) platforms demand immense compute acceleration and high-speed data throughput. However, system architects and AI administrators routinely hit a physical wall: PCIe slot starvation. In compact 1U, 2U, and edge server nodes, physical motherboard slots are scarce, rigid riser cards restrict airflow, and host CPUs lack sufficient lane allocation or native bifurcation support to expand peripherals seamlessly. HighPoint addresses these architectural constraints with an ecosystem of PCIe Gen 5 MCIO Expansion Solutions. By combining high-density MCIO (Mini Cool Edge IO) cabling with modular bridge cards and intelligent PCIe switching and Retimer adapters, administrators can route Gen 5 bandwidth anywhere within the existing chassis to add full-bandwidth PCIe slots, GPUs, DPUs, and NVMe arrays without reengineering the entire computing platform. HighPoint’s Approach to MCIO Expansion HighPoint’s expansion architecture is divided into three distinct hardware tiers to match different system topologies, BIOS capabilities, and density requirements. 1. Direct Onboard MCIO Expansion The Architecture: Motherboard Onboard MCIO 8i Ports + MCIO-PCIEX16-G5 Bridge Card How It Works: Modern server motherboards (AMD EPYC, Intel Xeon) and carrier boards feature native onboard MCIO 8i connectors. HighPoint’s MCIO-PCIEX16-G5 bridge card accepts dual MCIO 8i connections and converts them directly into a physical, full-length PCIe 5.0 x16 expansion slot. Primary Pain Point Addressed: Unused native board connectivity and severe slot scarcity. It provides a dedicated PCIe x16 slot without consuming any physical host PCIe slots on the motherboard. 2. Retimer / Bifurcation-Dependent Expansion The Architecture: Rocket 1624L (Internal MCIO) or Rocket 1624M (External MCIO) + MCIO-PCIEX16-G5 Bridge Card How It Works: This tier utilizes PCIe Gen 5 retimer adapters to extend host PCIe signals over internal or external cabling while maintaining signal integrity at 32 GT/s. The host slot relies on PCIe bifurcation (e.g., x16 bifurcated to x8/x8 or x4/x4/x4/x4) to split the bus across dual MCIO 8i ports. Primary Pain Point Addressed: Signal degradation over long cable runs and rigid mechanical layout limitations. It allows administrators to cost-effectively relocate high-power cards (like GPUs or NICs) to auxiliary bays or external enclosures with minimal latency. 3. Switch-Based Advanced MCIO Expansion The Architecture: Rocket 1628A (4x MCIO 8i) or Rocket 1624A (2x MCIO 8i) + MCIO-PCIEX16-G5 Bridge Card How It Works: These adapters utilize Broadcom PCIe Gen 5 switch ICs to create a dedicated, independent internal PCIe switching fabric. They do not require host motherboard bifurcation and run off native OS bus drivers. Administrators can link the MCIO ports to the MCIO-PCIEX16-G5 bridge card to create an extra PCIe x16 or dual x8 peripheral slots, or route links directly to high-density U.2/U.3/E3.S NVMe storage backplanes. Primary Pain Point Addressed: Bifurcation limitations, host CPU overhead, and peer-to-peer latency bottlenecks. The integrated switch enables Direct Peer-to-Peer (P2P) communication—allowing accelerators and storage to communicate across the switch fabric without loading the host CPU or traversing system RAM. Architectural Comparison Matrix Solution Tier Core Hardware Bifurcation Required? Host PCIe Slot Consumed? Max Bandwidth Key AI / Enterprise Capability Tier 1: Direct Onboard Onboard MCIO + MCIO-PCIEX16-G5 System Dependent No (Uses onboard ports) PCIe 5.0 x16 (64 GB/s) Zero PCIe host slot footprint; converts native board ports into GPU/NIC slots. Tier 2: Retimer / Bifurcation Rocket 1624L/M + MCIO-PCIEX16-G5 Yes (Host must support x8/x8) Yes (1x x16 Slot) PCIe 5.0 x16 (64 GB/s) Low-cost signal extension for internal/external remote card placement. Tier 3: Intelligent Switch Rocket 1624A / 1628A + MCIO-PCIEX16-G5 No (Built-in Switch Engine) Yes (1x x16 Slot) PCIe 5.0 x16 (64 GB/s aggregate) Direct P2P GPU/NIC offload, concurrent NVMe + compute expansion, universal motherboard support. Choosing the right Tier Select Tier 1 - Direct Onboard connection (MCIO + MCIO-PCIEX16-G5 add up to 2x PCIe x8 Slots or a single x16 Slot): when your server motherboard already features unpopulated MCIO 8i connectors and you need an extra GPU, DPU, or capture card without consuming existing PCIe expansion slots. Select Tier 2 - Retimer-Based (Rocket 1624L/1624M- add up to 2x PCIe x8 Slots or a single x16 Slot): when running platforms that fully support PCIe bifurcation, and your primary goal is clean internal thermal layout or external chassis extension at minimal component cost. Select Tier 3 - Switch-Based (Rocket1624A/1628A – add up to 4x PCIe x8 slots or 2x PCIe x16 Slots): when designing high-throughput AI inference nodes, HPC clusters, or mixed compute/storage servers. This is mandatory for systems lacking robust BIOS bifurcation, environments requiring direct P2P device communication, or setups multiplexing multiple SSD arrays alongside dedicated accelerators. Learn More

The PCIe Gen5 era has fundamentally changed the power requirements for high-performance computing. While many external enclosures rely on standard power supplies that deliver between 400 and 850W, this range of capacity is no longer sufficient for today's flagship GPUs and accelerator cards. For mission-critical AI training, simulations and other HPC workflows, the only way to ensure unconstrained performance is with the dedicated 1300W PSU support found in the RocketStor 8631D. The Hidden Danger: Transient Power Spikes The maximum TDP (Thermal Design Power) rating of a high-end accelerator might be presented as 600W. However, the true challenge lies in transient power spikes. During intense computational phases—the precise moments when your device is pushing its limit—it can draw significantly more power for a few milliseconds. If your power supply unit (PSU) is undersized, these transient spikes can lead to a variety of critical issues: 1. Voltage Sag: If the device attempts to draw more power than what the PSU can provide, a voltage dip or sag will occur, resulting in instability and performance loss. 2. Performance Throttling: Repeated power sags can result in the device’s firmware aggressively throttling its output in order to prevent a crash, leading to unpredictable, non-inconsistent operation. 3. System Instability: In extreme cases, the enclosure may shut down, corrupting ongoing training runs or simulations. Why 850W is often Not Enough for the Latest Accelerators Cards An 850W enclosure PSU must power the GPU and the entire enclosure infrastructure; cooling fans, integrated chipset (such as a retimer or switch), and the Host Interface Card overhead). For a 600W accelerators, an 850W PSU leaves 250W of head room. This limited margin is easily consumed by transient spikes and cooling demands, forcing the GPU to throttle its performance precisely when maximum TDP is needed. The Solution: RocketStor 8631D - Over-Engineered Power Delivery The RocketStor 8631D is engineered specifically for this power challenge. By integrating a dedicated 1300W PSU, the enclosure provides a massive power buffer, ensuring: Sustained TDP: A 600W accelerator card still leaves 700W of overhead for spikes, cooling, and reliability; more than double its standard requirement! Zero Throttling: The accelerator maintains its maximum boost clocks and performance profile without risk of a voltage sag or performance throttling. Unmatched Longevity: Stable, high-quality power delivery protects your expensive Gen5 accelerators from unnecessary stress. Learn More: Rocket 8631D PCIe Gen5 x16 External CopprLink Expansion Enclosure Signal Integrity Solved: Why Astera Labs Gen5 Retimers are Mandatory for External x16 Bandwidth The Ultimate Gen5 Enclosure Checklist: Why the 8631D Outperforms Market Alternatives