SYSTEM
Scan code: LN106116 Manufacturer code: DLTT880X2

NVIDIA Tesla GPU accelerated server for Deep Learning training

3XS DLT T880X2, 8x 32GB Tesla V100S, 2x Xeon Gold 6248, 512GB DDR4, 1TB NVMe SSD, 3.84TB SSD, Ubuntu 18.04

Thumbnail 1 : NVIDIA Tesla GPU accelerated server for Deep Learning training Thumbnail 2 : NVIDIA Tesla GPU accelerated server for Deep Learning training Thumbnail 3 : NVIDIA Tesla GPU accelerated server for Deep Learning training Thumbnail 4 : NVIDIA Tesla GPU accelerated server for Deep Learning training
Scan code: LN106116 Manufacturer code: DLTT880X2
£90,999.98
Order now and get it tomorrow (upgrade to Sunday delivery at checkout)
From £2,485.33 / month (APR 14.9%)
Monthly Payment: £8,169.34
Instalment Rate: 14.9% p.a. (fixed)
Cost of purchase: £90,999.98
Total inc. interest: £98,031.99
Monthly Payment: £4,367.88
Instalment Rate: 14.9% p.a. (fixed)
Cost of purchase: £90,999.98
Total inc. interest: £104,828.96
Monthly Payment: £3,108.83
Instalment Rate: 14.9% p.a. (fixed)
Cost of purchase: £90,999.98
Total inc. interest: £111,917.82
Monthly Payment: £2,485.33
Instalment Rate: 14.9% p.a. (fixed)
Cost of purchase: £90,999.98
Total inc. interest: £119,295.89
If you are approved for a credit limit with PayPal Credit and use it for future purchases, the APR for those purchases won't be more than 19.9% and may be even lower.

Representative Example:

Purchase Rate: 19.9% p.a. (variable)

Representative: 19.9% APR (variable)

Assumed Credit Limit: £1,200

Subject to status. Terms and Conditions apply.
Delivery Options
  • By DPD On 17th Aug to your specified address. | £11.50 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 window

    Weekend, timed and European delivery options are available at checkout
  • Collect instore Collect from our Bolton store, BL6 6PE | Free

    Order online, collect from our Bolton store (25-28 Enterprise Park, Middlebrook, Horwich, Bolton, BL6 6PE)

Scan 3XS Deep Learning Training Servers

NVIDIA GPU accelerators provide fast and accurate training of neural networks and other AI analytics workloads. Scan offers a range of datacentre servers powered by the latest Turing-architecture enterprise-class NVIDIA Tesla GPU accelerators, providing you with maximum performance and reliability.

This Deep Learning Training Server features eight 32GB NVIDIA Tesla V100S GPU accelerators with pair of Intel Xeon Gold host CPUs plus 512GB of DDR4 memory. A super-fast 1TB NVMe SSD is provided for the OS and caching plus 3.84TB SATA SSD for larger datasets.

The system is soak tested with Deep Learning workloads and is pre-installed with the Ubuntu 18.04 operating system plus a custom software stack built on NVIDIA CUDA that includes Docker-CE, Nvidia-Docker2 and GPU-optimised libraries.

docker
tensorflow

Scan Deep Learning Training Servers

Scan is an Elite Solution Provider for NVIDIA DGX Systems and has a dedicated AI team including data scientists to our support our enterprise-class Deep Learning Training Servers.

Detailed Specifications

Component Description
GPU Accelerator 8 x 32GB NVIDIA Tesla V100S
Processor 2 x Intel Xeon Gold 6248 with 20 Cores / 40 Threads
Memory 512GB ECC Registered DDR4
System Drive 1TB Samsung 970 Evo Plus NVMe SSD
Storage Drive 3.84TB Intel D3-S4510 SSD
Networking 2 x Intel 1GbE LAN plus a dedicated 1GbE NIC for IPMI remote management
Base Unit Asus ESC8000 G4
Form Factor 4U, 798mm deep
Operating System Ubuntu 18.04
Warranty 3 Year Premium Warranty - 1st Year Onsite, 2nd & 3rd Year Return to Base (inc Parts & Labour)

Single or Dual Root Topology

The 3XS DLT Q880X2 features a switchable topology enabling you to select the optimum configuration for your current workload.

SINGLE ROOT CONFIGURATION

Single Root ConfigurationChoosing a single root configuration enables all the GPUs communicate with one another via a single CPU, reducing latency for data shared between the GPUs.

DUAL ROOT CONFIGURATION

Dual Root ConfigurationChoosing a dual root configuration splits the eight GPUs into two sets of four, each set of GPUs communicating with one of the CPUs, increasing performance in CPU-heavy workloads.

Why buy a 3XS Deep Learning Training Server?

ENTERPRISE GRADE SERVERS

Our goal is to make the most reliable servers, using the best quality components, backed by our award-winning customer support. 3XS Systems builds thousands of PCs, laptops, workstations, servers and storage devices a month in our state-of-the-art factory in the UK, and has done for more than 10 years. Our reliable business IT solutions are trusted and used by companies of all sizes plus government and education and our process and quality control systems conform to ISO:9001 standards.

No Compatibility Headaches

It can take weeks to research the right combination of components, not just to ensure you get the most bang for your buck but to minimise the chance of one component being incompatible with another. This is especially true for Deep Learning and AI as these workloads put incredible load on components such as the GPU, motherboard and power supply. We specially choose all the components in our Deep Learning Training Servers with these conditions in mind, ensuring your system works flawlessly and delivers consistent performance.

Intensively Tested

Building a server only takes a few hours, but installing all the software and making sure that everything is configured properly and that your software runs stably takes up a huge amount of time. By ordering a 3XS Deep Learning Training Server our engineers take this painful process away, as all our systems undergo a rigorous Quality Control procedure which includes installing Ubuntu 18.04, installing the latest drivers, BIOS. Deep Learning Training Servers also have to pass an intensive stress test including real Deep Learning frameworks such as TensorFlow.

Peace of Mind

Unfortunately server components do occasionally fail, and when this happens if you’ve built your system yourself you’re on your own. However, with a professionally built 3XS Deep Learning Training Server we have a dedicated team of support engineers and data scientists to help resolve any issues. Our technical support team is available 7 days a week, from 9am to 10pm Mon - Fri, 10am to 6pm on Saturdays and 10am to 4pm on Sundays. Click here for more info on how to get technical support for your 3XS system.

If our support team is unable to fix your Deep Learning Training Server remotely and it is still under warranty we’ll either send out an engineer to visit your datacentre or office to repair it or have it repaired in our state of the art factory in Lancashire.

Next Day Delivery

3XS Deep Learning Training Servers are pre-built, soak tested and ready to ship from our warehouse for next-day delivery in the mainland UK.

Warranty & Technical Support

3XS Deep Learning Training Servers are covered by our Premium 3 Year Warranty. This means that if your system develops a fault in its first year we can send an engineer to your home or office to repair it. In the remaining two years of the warranty we'll still repair or replace any faulty parts, but do this from our state of the art factory in Lancashire.
Click here for more info on how to get technical support for your 3XS system.

Specifications
Edition DLT T880X2
Form Factor 4U Rackmount
CPU Included Yes
CPU Type 2x Intel Xeon Gold
CPU Model Intel Xeon Gold 6248
Architecture Cascade Lake
Manufacturing Process 14 nm
Unlocked CPU No
Core 20 Core
No. of Threads 40
Clock Speed 2.5 GHz
Burst Frequency 3.9 GHz
CPU Cache 27.5MB
CPU Cooler Passive Heatsink
Max. TDP 150 W
CPU Socket 3647 x2
Chipset Intel C621
Motherboard Form Factor Bespoke/Custom
Motherboard Model  
Memory Type DDR4
Memory Included Yes
Memory Capacity Included 512GB (16x32GB)
Memory Edition  
Memory Channels Hex (6) per CPU/Socket
Memory Type (ECC) ECC
Memory Type (R/U) RDIMM (Registered)
Memory Speed (Mhz) DDR4 - 2666
Max. Memory Capacity 3072GB (24x128GB)
On-board/Integrated Graphics Yes
Processor/Motherboard Graphics Chip Aspeed AST2500 with 64MB VRAM
Discrete GPU Yes
Discrete GPU Chipset 8x NVIDIA Tesla V100S
GPU Architecture Volta
GPU Manufacturing Process 12 nm
GPU Cores/Streams/Execution Units 40960 (8x 5120)
GPU Tensor Cores N/A
Base GPU Speed N/A
Max. GPU Speed N/A
GPU Memory Size 32GB
GPU Memory Type HBM2
GPU Memory Clock  
GPU Memory Bus  
Discrete GPU Cooler Air Cooled
Multi-GPU Configuration Yes
Max. GPU Resolution  
Graphics Interface PCIe (x16)
Expansion Slots N/A
SATA SATA3 / M.2
HDD Storage Included Yes
HDD Storage Type 1x M.2 SSD + 1x 2.5" SATA SSD
HDD Storage Capacity 1TB
Optical Drive N/A
Drive Bays 8 x 2.5" SATA Hot-Swap
Backplane I/O  
RAID Support Intel® SATA RAID 0/1/10/5
Internal I/O
  • 2 x M.2 Connector
  • 8 x SATA III - 6Gb/s
  • 1 x Serial (COM) Header
  • 2 x USB 2.0 Port
  • 2 x USB 3.0 Connector
  • 1 x VGA connector
Network Interface Type 4x EDR Infiniband
Network Chip/Modules  
Front Panel
  • 1 x HDD Activity LED
  • 2 x LAN Activity LEDs
  • 1 x Message LED
  • 1 x Power Switch with LED
Back Panel
  • 1 x HDD Activity LED
  • 1 x Location Switch with LED
  • 1 x Message LED
  • 1 x Power Switch with LED
  • 1 x Q-Code/Port 80 LED
  • 1 x RJ-45 Mgmt LAN
  • 2 x RJ45 Dedicated IPMI port
Cooling Fans  
Keyboard / Mouse Included No
Remote Control Included No
VESA Mountable No
PSU Wattage  
PSU Efficiency Rating  
PSU Redundancy N/A
Operating System Included Yes
Operating System Ubuntu Linux Host OS
Colour Grey/ Silver
Dimensions 444 x 131 x 866 (mm WxHxD)
Weight 60.78 kg
Includes  
Notes  
Storage Drives
SSD Storage Included No
SSD Storage Type  
SSD Storage Capacity  
Additional Information
Scan Code LN106116
Model Number DLTT880X2
GTIN 5055948917724
Warranty

Please note your statutory rights are not affected.

For further information regarding Scan's warranty procedure please see our terms and conditions

Details
Duration:
36 months
Type:
Return to base
DOA Period:
7 days
RTB Period:
36 months
Manufacturer Contact Details
Telephone:
01204 474756
TekSpek Guides
NVIDIA DGX-2 Supercomputer
NVIDIA DGX-2 Supercomputer
Date Issued: 3rd Sep 2018

The increase in processing speed for computers has opened up new avenues of research and provided deeper insights into solutions for complex problems. Such power has been harnessed to accelerate research into medicine, better predict weather patterns, perform hugely complex calculations for the oil and gas industry, and a whole host more.

Read More

NVIDIA DGX-1 Supercomputer
NVIDIA DGX-1 Supercomputer
Date Issued: 21st Jul 2017

The increase in processing speed for computers has opened up new avenues of research and provided deeper insights into solutions for complex problems. Such power has been harnessed to accelerate research into medicine, better predict weather patterns, perform hugely complex calculations for the oil and gas industry, and a whole host more.

Read More

Reviews
Questions & Answers
NVIDIA Tesla GPU accelerated server for Deep Learning training
NVIDIA Tesla GPU accelerated server for Deep Learning training