Adaptable Accelerator Cards for Data Center Workloads
MA Labs provides all kinds of hardware solutions to Data Centers and now is proudly presenting the Xilinx Data Center Acceleration cards. These acceleration cards offer up to 90x higher performance than CPUs on key workloads at one-third the cost and over 4x higher inference throughput and 3x latency advantage over GPU-based solutions. Its adaptability is wide and as workload algorithms evolve, it can use reconfigurable hardware to adapt faster than fixed-function accelerator card product cycles. You can also use these products to deploy solutions in the cloud or on-premises interchangeably, scalable to application requirements.
If you are a B2B customer, please email us at email@example.com. Our dedicated sales team will contact you shortly.
This low-profile, PCIe-based media accelerator card can deliver a high-density real-time transcoding solution for live streaming video service providers, OEMs and Content Delivery Network.Datasheet >
Delivers compute, networking, and storage acceleration in an efficient 75-watt, small form factor, and armed with 100 GbE networking, PCIe Gen4, and HBM2. Designed to deploy in any server.Datasheet >
Incredible compute, networking, and storage acceleration thanks to 890k LUTs, 5.9k DSP slices, 64GB of DDR4 memory, and dual 100Gbps network interfaces.Datasheet >
Offers 1.3M LUTs, 11.5k DSP slices, 64GB of DDR4 memory, dual 100Gbps network interfaces, and delivers 90x higher performance than CPUs on key workloads at a fraction of the costs.Datasheet >
Built for compute and memory bound workloads and is armed with 8GB of HBM2 + 32GB of DDR4 memory, 1.1M LUTs , 8.5k DSP slices, dual 100Gbps network interfaces, PCIe Gen4, and support for CCIX.Datasheet >
|Supported Alveo card||- A Xilinx Alveo-qualified server support for Alveo U50 to provide optimized acceleration for workloads across cloud and on-premises data centers||- A Xilinx Alveo-qualified server support for Alveo U280 to provide optimized acceleration for workloads across cloud and on-premises data centers||- A Xilinx Alveo-qualified server support for Alveo U200 to provide optimized acceleration for workloads across cloud and on-premises data centers|
|Processor||- Powered by AMD EPYC™ 7002 processor with 64 cores, 128 threads, and support for max. TDP of up to 280 watts per socket||- Dual AMD EPYC™ 7002 series processor family||- Dual Socket P (LGA 3647) 2nd Gen Intel® Xeon® Scalable Processors and Intel® Xeon® Scalable Processors‡, Dual UPI up to 10.4GT/s, Support CPU TDP 70-205W|
|Supported GPU Type||- GPU-optimized design allows four double-slot or eight single-slot GPUs, including NVIDIA® A100, Tesla® T4 or Quadro®||- Supports up to 3 x double slot GPU cards. NVIDIA® NGC™ Ready server||- Support 1 x Xilinx U200, keep temperature at 30C with additional fan|
|Memory||- 8-Channel Maximum up to 2048GB RDIMM DDR4 3200, 8 x DIMMs||- 8-Channel RDIMM/LRDIMM DDR4 per processor, 32 x DIMMs||- 24 DIMM slots, up to 6TB 3DS ECC DDR4- 2933MHz† RDIMM/LRDIMM, supports Intel® Optane™ DCPMM††|
|Storage||- 8 x 3.5" or 2.5" Hot-swap Storage Bays||- 12 x 3.5" SATA hot-swappable HDD/SSD bays||- 24 Hot-swap 2.5" drive bays: 24 SAS3 via opt. AOC (4 hybrid ports - opt. 4 NVMe)|
|Expansion Slots||- 4 x PCIe x16 slots (Gen4 x16 link, FH,FL) for dual-slot GPU cards or 8 x PCIe x16 slots (Gen4 x8 link, FH,FL) for single-slot GPU cards||- 5 x PCIe Gen4 x16 expansion slots||- 1 PCI-E 3.0 x16 (FH, 10.5" L), 5 PCI-E 3.0 x8 (FH, 10.5" L), 1 PCI-E 3.0 x8 (LP) and 1 PCI-E 3.0 x8 (internal LP)|
|Power supply||- 1+1 Redundant 2200W 80 PLUS Platinum Power Supply||- 2000W 80 PLUS Platinum redundant power supply||- 1000W Redundant Power Supplies Titanium Level (96%)|
|Warranty||- 3 years manufacturer warranty with advanced replacement||- 3 years manufacturer warranty with advanced replacement||- 1 year manufacturer warranty|
If you order server systems from us, our engineers will provide professional system integration and testing services before shipping and our dedicated Server Solution Team will provide you world-class before/after-sales support.