NVIDIA ConnectX-5 MCX555A-ECAT InfiniBand Adapter Card
Single-Port QSFP28 | 100Gb/s InfiniBand & Ethernet | PCIe 3.0 x16 | RDMA Enabled
High-performance, low-latency 100Gb/s network adapter designed for HPC, AI, and cloud data centers. Featuring advanced offloads including NVMe over Fabrics, GPUDirect RDMA, and tag matching for MPI workloads — delivering industry-leading throughput and CPU efficiency.
Product Overview
The NVIDIA ConnectX-5 MCX555A-ECAT is a single-port 100Gb/s InfiniBand adapter card in a low-profile PCIe form factor. Leveraging the proven ConnectX-5 architecture, it delivers up to 100Gb/s throughput with sub-microsecond latency and a high message rate. The card supports both InfiniBand (up to EDR) and 100GbE, providing versatile connectivity for high-performance computing, storage, and virtualized environments.
Built with an embedded PCIe switch and advanced RDMA capabilities, the MCX555A-ECAT offloads critical communication tasks from the CPU — enabling higher application performance, lower power consumption, and reduced total cost of ownership. It is fully compatible with PCIe 3.0 x16 slots and supports a wide range of operating systems and acceleration frameworks.
Key Features
- Up to 100Gb/s connectivity per port (InfiniBand EDR / 100GbE)
- Single QSFP28 connector for optical or copper cables
- PCIe 3.0 x16 host interface (auto-negotiates to x8, x4, x2, x1)
- RDMA, send/receive semantics with hardware-based reliable transport
- Tag matching and rendezvous offloads for MPI and SHMEM
- NVMe over Fabrics (NVMe-oF) target offloads for efficient storage
- GPUDirect RDMA (PeerDirect) acceleration for GPU communication
- Hardware-based congestion control & adaptive routing support
- SR-IOV virtualization: up to 512 virtual functions
- RoHS compliant, low-profile form factor (tall bracket included)
Advanced Technology & Offloads
The ConnectX-5 architecture integrates a range of hardware acceleration engines that reduce CPU intervention and improve application scalability:
- MPI Tag Matching & Rendezvous Offload: Offloads message matching and rendezvous protocol processing, dramatically improving MPI performance for HPC clusters.
- Out-of-Order RDMA with Adaptive Routing: Enables efficient use of multiple network paths while maintaining ordered completion semantics, maximizing fabric utilization.
- NVMe-oF Target Offloads: Allows NVMe storage systems to serve remote access with near-zero CPU overhead, ideal for disaggregated storage architectures.
- Dynamically Connected Transport (DCT): Provides extreme scalability for large compute and storage systems by eliminating connection setup overhead.
- ASAP2 Accelerated Switching & Packet Processing: Hardware offload for Open vSwitch (OVS) and overlay network tunneling (VXLAN, NVGRE, GENEVE).
- On-Demand Paging (ODP): Supports virtual memory paging for RDMA operations, simplifying application development.
Typical Deployments
- High-Performance Computing (HPC): Ideal for supercomputing clusters, MPI-based simulations, and scientific research workloads requiring low latency and high message rates.
- AI & Deep Learning Training: Combined with GPUDirect RDMA, enables fast GPU-to-GPU communication across nodes, accelerating training times.
- NVMe-oF Storage Systems: Deploy as storage targets or initiators in NVMe over Fabrics environments for high-throughput, low-latency block storage access.
- Cloud & Virtualized Data Centers: SR-IOV and virtualization offloads support multi-tenant environments with guaranteed QoS and secure isolation.
- High-Frequency Trading (HFT): Ultra-low latency and hardware timestamping (IEEE 1588v2) meet the demands of financial services applications.
Compatibility & Interoperability
The MCX555A-ECAT is designed for broad compatibility with NVIDIA InfiniBand switches (e.g., Quantum, Spectrum) and third-party 100GbE switches. It supports both passive copper DACs and active optical cables via QSFP28 ports.
Operating Systems & Software Stacks:
- RHEL / CentOS, Ubuntu, Windows Server, FreeBSD, VMware ESXi
- OpenFabrics Enterprise Distribution (OFED) / WinOF-2
- NVIDIA HPC-X, OpenMPI, MVAPICH2, Intel MPI, Platform MPI
- Data Plane Development Kit (DPDK) for kernel bypass
Technical Specifications
| Parameter |
Specification |
| Model |
MCX555A-ECAT |
| Form Factor |
PCIe Low-Profile (14.2cm x 6.9cm without bracket), Tall bracket pre-installed, short bracket included |
| Port Speed & Type |
1x QSFP28, up to 100Gb/s InfiniBand (EDR) and 100GbE |
| Host Interface |
PCI Express 3.0 x16 (compatible with x8, x4, x2, x1) |
| InfiniBand Support |
IBTA 1.3 compliant, 100Gb/s EDR, FDR, QDR, DDR, SDR; 8 virtual lanes + VL15; 16 million I/O channels |
| Ethernet Support |
100GbE, 50GbE, 40GbE, 25GbE, 10GbE, 1GbE; IEEE 802.3cd, 802.3bj, 802.3by, 802.3ba, 802.3ae |
| RDMA Capabilities |
RDMA over Converged Ethernet (RoCE), hardware reliable transport, out-of-order RDMA, atomic operations |
| Storage Offloads |
NVMe over Fabrics target offload, iSER, SRP, NFS RDMA, SMB Direct, T10 DIF signature handover |
| Virtualization |
SR-IOV (up to 512 virtual functions), VMware NetQueue, NPAR, PCIe Access Control Services (ACS) |
| CPU Offloads |
TCP/UDP/IP stateless offload, LSO/LRO, checksum offload, RSS/TSS, VLAN/MPLS tag insertion/stripping |
| Overlay Networks |
Hardware offload for VXLAN, NVGRE, GENEVE encapsulation/decapsulation |
| Management |
NC-SI over MCTP, PLDM for monitor/control and firmware update, I2C, SPI, JTAG |
| Remote Boot |
Remote boot over InfiniBand, Ethernet, iSCSI; UEFI, PXE support |
| Power Consumption |
Not publicly specified; typical sub-20W range – please confirm for your system |
| Operating Temperature |
0°C to 55°C (typical environment) |
| Compliance |
RoHS, REACH, FCC, CE, VCCI, ICES, RCM |
Note: Specifications derived from NVIDIA ConnectX-5 product documentation. For the latest details and firmware support, refer to official NVIDIA release notes.
Selection Guide – ConnectX-5 Family
| Ordering Part Number |
Ports / Speed |
Host Interface |
Form Factor |
Key Features |
| MCX555A-ECAT |
1x QSFP28, 100Gb/s |
PCIe 3.0 x16 |
Low-profile PCIe |
Standard single-port, EDR InfiniBand / 100GbE |
| MCX556A-ECAT |
2x QSFP28, 100Gb/s |
PCIe 3.0 x16 |
Low-profile PCIe |
Dual-port, EDR/100GbE |
| MCX556A-EDAT |
2x QSFP28, 100Gb/s |
PCIe 4.0 x16 |
Low-profile PCIe |
ConnectX-5 Ex, enhanced PCIe Gen4 |
| MCX556M-ECAT-S25 |
2x QSFP28, 100Gb/s |
2x PCIe 3.0 x8 |
Socket Direct |
Dual-socket server connection via harness |
| MCX545B-ECAN |
1x QSFP28, 100Gb/s |
PCIe 3.0 x16 |
OCP 2.0 Type 1 |
Open Compute Project form factor |
For OCP or Multi-Host variants, please contact sales. All cards support backward compatibility to lower speeds.
Why Choose ConnectX-5 MCX555A-ECAT
- Superior Application Performance: Hardware offloads for MPI, NVMe-oF, and overlays free CPU cores for business logic.
- Scalable RDMA Fabric: DCT, XRC, and out-of-order RDMA deliver linear scalability for thousands of nodes.
- GPU Acceleration Ready: GPUDirect RDMA enables direct memory access between GPUs and network adapters, eliminating CPU bottlenecks in AI clusters.
- Flexible Deployment: Single QSFP28 port simplifies cabling and is ideal for 100Gb/s leaf-spine architectures.
- Investment Protection: Support for both InfiniBand and Ethernet allows seamless transition between protocols as needs evolve.
Service & Support
Hong Kong Starsurge Group provides complete lifecycle support for NVIDIA ConnectX-5 adapters, including pre-sales configuration assistance, firmware update guidance, and warranty service. Our technical team can help with:
- Compatibility verification with your server and switch infrastructure
- Performance tuning for HPC or storage workloads
- Custom bracket options and bulk packaging requirements
- RMA processing and advanced replacement services
Contact our sales engineers for volume pricing and lead time information.
Frequently Asked Questions
Q: What is the difference between MCX555A-ECAT and MCX556A-ECAT?
A: MCX555A-ECAT has a single QSFP28 port, while MCX556A-ECAT has dual ports. Both support 100Gb/s per port and PCIe 3.0 x16. Choose single-port for simpler cabling or dual-port for higher density.
Q: Can this card be used in a PCIe 3.0 x8 slot?
A: Yes, the card auto-negotiates to x8, x4, x2, or x1 lane widths, though maximum throughput may be limited by available bandwidth.
Q: Does it support RoCE (RDMA over Converged Ethernet)?
A: Yes, ConnectX-5 supports RoCE for Ethernet fabrics, providing low-latency RDMA services on standard Ethernet networks.
Q: What cables are compatible?
A: The card works with QSFP28 passive copper DACs (up to 5m) and active optical cables (AOC) for longer reaches, as well as optical transceivers for fiber connectivity.
Q: Is this card supported in VMware ESXi?
A: Yes, VMware ESXi is supported with native drivers and SR-IOV capabilities. Please check the VMware compatibility guide for specific versions.
Handling & Installation Precautions
Electrostatic Discharge (ESD): Always use ESD-safe practices when handling the adapter. Store in anti-static packaging until installation. Cooling Requirements: Ensure adequate airflow in the server chassis to maintain operating temperature within specified range. Firmware Updates: Use NVIDIA official firmware tools (MFT) and verify compatibility with your OS and driver version before updating. Cable Bending: Follow QSFP28 cable bend radius guidelines to avoid signal degradation.
This is a Class A product. In a residential environment it may cause radio interference. Ensure proper shielding and grounding per local regulations.
About Hong Kong Starsurge Group Co., Limited
Founded in 2008, Hong Kong Starsurge Group Co., Limited is a technology-driven provider of network hardware, IT services, and system integration solutions. Serving customers worldwide with products including network switches, NICs, wireless access points, controllers, and high-speed cabling, Starsurge combines deep technical expertise with a customer-first approach. The company supports industries such as government, healthcare, manufacturing, education, finance, and enterprise, offering IoT solutions, network management systems, custom software development, and multilingual global delivery. With a focus on reliable quality and responsive service, Starsurge helps clients build efficient, scalable, and dependable network infrastructure.
Key Facts – NVIDIA MCX555A-ECAT
100Gb/s
Port speed (InfiniBand & Ethernet)
PCIe 3.0 x16
Host interface bandwidth
512 VFs
Maximum SR-IOV virtual functions
NVMe-oF
Hardware offloaded storage target
Compatibility Matrix
| Component / System |
Compatibility Status |
Notes |
| NVIDIA Quantum InfiniBand Switches |
Certified |
EDR, HDR compatibility when using appropriate firmware |
| NVIDIA Spectrum Ethernet Switches |
Certified |
100GbE, 50GbE, 25GbE modes supported |
| Third-party 100GbE switches |
Compatible |
Requires IEEE standards compliance; tested with major vendors |
| GPU servers (NVIDIA DGX, HGX) |
Certified with GPUDirect |
RDMA acceleration for multi-GPU communication |
| Storage arrays with NVMe-oF |
Supported |
Target offload enables efficient NVMe fabric access |
Buyer Checklist – 100Gb/s InfiniBand Adapter
- ☑ Confirm server has an available PCIe 3.0 x16 (or x8) slot with adequate clearance.
- ☑ Determine port count: single-port (MCX555A-ECAT) vs dual-port (MCX556A-ECAT).
- ☑ Choose cable type: passive copper DAC for short distances (≤5m) or optical for longer runs.
- ☑ Verify operating system and driver support (OFED, Windows, VMware).
- ☑ For GPU clusters, ensure GPUDirect RDMA compatibility with your GPU model and driver version.
- ☑ Check if tall or short bracket is required for your server chassis.
Related Products
- NVIDIA MCX556A-ECAT – Dual-port 100Gb/s ConnectX-5 adapter
- NVIDIA MCX556A-EDAT – ConnectX-5 Ex with PCIe 4.0 support
- NVIDIA Quantum-2 QM9700 40-port 800Gb/s InfiniBand Switch
- Mellanox QSFP28 passive DAC cables (1m, 2m, 3m)
- NVIDIA Spectrum-4 SN5600 100GbE/400GbE Ethernet switches
Related Guides & Resources
- NVIDIA ConnectX-5 InfiniBand Adapter Card User Manual
- RDMA over Converged Ethernet (RoCE) Deployment Guide
- GPUDirect RDMA Best Practices for AI Clusters
- NVMe over Fabrics with ConnectX-5 – Configuration Guide
- OFED Installation and Tuning Guide