96TB NVMe over Fabric Storage Appliance
96TB DLS4000 Deep Learning Storage Appliance
to your specified address. |
Receive SMS with one-hour delivery window
Weekend, timed and European delivery options are available at checkout
Have your parcel delivered by DPD to your specified address. Receive SMS with one-hour delivery windowWeekend, timed and European delivery options are available at checkout
Collect from our Bolton store, BL6 6PE |
Order online, collect from our Bolton store (25-28 Enterprise Park, Middlebrook, Horwich, Bolton, BL6 6PE)
Deep Learning Storage Appliances
A.I. Optimised Solutions, made by A.I. experts
High performance deep learning training solutions from the NVIDIA DGX family require low-latency storage solutions. This 2U NVMe over Fabric Storage appliance is powered by Excelero and although starts at 32TB, can scale up to 330TB of storage within a single enclosure.
The starting configuration of 4 SSDs is designed for the DGX user who’s training data has grown larger than internal storage and wants to minimise training time. With performance that will ensure that the DGX-1 GPUs run at full efficiency with approximate 6500 FPS (RESnet50) or over 2M IOPS, 12GB/s random read with less than 90 microseconds latency. Each appliance is scalable up to 22 SSDs, providing more than 23GB/sec from a single enclosure.
The solution scales-out, meaning that as you add additional 2Us appliances the capacity can be consolidated as a whole solution and the performance is doubled. For example, two appliances with 160TB total capacity would provide 46GB/sec performance.
NVMe over Fabric Low Latency Storage
Excelero delivers low-latency distributed block storage for web-scale applications. NVMesh enables shared NVMe across any network and supports any local or distributed file system. The solution features an intelligent management layer that abstracts underlying hardware with CPU offload, creates logical volumes with redundancy, and provides centralised, intelligent management and monitoring.
Applications can enjoy the latency, throughput and IOPs of a local NVMe device with the convenience of centralised storage while avoiding proprietary hardware lock-in and reducing the overall storage TCO. NVMesh features a distributed block layer that allows unmodified applications to utilize pooled NVMe storage devices across a network at local speeds and latencies. Distributed NVMe storage resources are pooled with the ability to create arbitrary, dynamic block volumes that can be utilised by any host running the NVMesh block client.
Intel underpins the infrastructures of the world’s leading Enterprises and bring the same level of commitment to quality, performance and endurance to the Scan 3XS Deep Learning Storage Appliances.
Based upon Intel’s proven NVMe SSD and server hardware, the appliance provides the level of maturity needed for such high performance ultra-low latency solutions.
For AI optimised storage, maximum throughput and low latency across the network are key.
We have partnered with Mellanox to provide our 3XS Storage Appliances with 100Gb/s Infiniband cards which are ideally suited to connecting to the NVIDIA DGX range of compute appliances.
|Storage Capacity||96TB per node|
|NVMesh license||Perpetual, including unlimited clients|
|Connectivity||2 x QSP28 EDR Inifinband /100Gb/s Ethernet|
|Power||Hot-swap redundant PSUs|
|Form Factor||2U, 710mm deep|
|Installation & Configuration||Service packages range depending on individual requirements|
|Warranty & Support||Software 1/3 year Platinum Support Package, 24x7, 365 Days. Hardware options to suit the individual requirements|
|CPU Type||2nd Gen. Intel Xeon Scalable Processors|
|Architecture||Cascade Lake-SP (Purley)|
|No. of Threads||N/A|
|Motherboard Form Factor|
|Memory Type||DDR4 ECC Registered|
|Memory Capacity Included||96GB|
|Memory Type (ECC)||ECC|
|Memory Type (R/U)||RDIMM (Registered)|
|Memory Speed (Mhz)|
|Max. Memory Capacity|
|Processor/Motherboard Graphics Chip||N/A|
|Discrete GPU Chipset|
|GPU Manufacturing Process|
|GPU Cores/Streams/Execution Units|
|GPU Tensor Cores||N/A|
|Base GPU Speed||N/A|
|Max. GPU Speed||N/A|
|GPU Memory Size|
|GPU Memory Type|
|GPU Memory Clock|
|GPU Memory Bus|
|Discrete GPU Cooler|
|Max. GPU Resolution|
|Storage Drive Included||Yes|
|Storage Drive Type|
|Storage Drive Capacity||96TB|
|Network Interface Type||2x QSFP28 100GbE InfiniBand|
|Keyboard / Mouse Included||No|
|Remote Control Included||No|
|PSU Efficiency Rating|
|Operating System Included||No|
|Dimensions||438 x 87.3 x 704.9 (mm WxHxD)|
Please note your statutory rights are not affected.
For further information regarding Scan's warranty procedure please see our terms and conditions
- 36 months
- Return to base
- DOA Period:
- 7 days
- RTB Period:
- 36 months
- 01204 474756