We take cryptocurrency — so the coins you hold can buy real hardware.Pay in crypto, spend it on real hardware. Free insured shipping worldwide over $399. Plain, unbranded boxes, sent from Germany.Free insured shipping over $399. Your rate is locked for 30 minutes once checkout opens.Rate locked 30 minutes. Every item sealed, serial-checked and covered by a 24-month warranty.Sealed, 24-month warranty.

Saved items Sign in
HP Tesla M2075 GPU card with 6 GB GDDR5 memory

HPE

HP M2075GPU Tesla graphics card 6 GB GDDR5

SKU CH-9Z6S-DJQRB MPN M2075GPU

Adds 6 GB GDDR5 GPU compute with 655 double-precision Gflops and 1.3 single-precision Tflops to ProLiant servers

  • 6 GB GDDR5 memory with ECC for large dataset handling
  • 3 Tflop single-precision peak for accelerated compute workloads
  • 655 Gflop double-precision peak for scientific accuracy
  • CUDA programming interface enables GPU-aware cluster scheduling
  • Designed for integration with HP ProLiant servers

Server GPU computing module

The HP M2075GPU is a Tesla graphics card built for GPU computing inside HP ProLiant servers. It provides 6 GB of GDDR5 memory with ECC for data integrity. The board delivers 655 double-precision Gflops and over 1.3 single-precision Tflops. It is intended for scale-out deployments that run CUDA workloads.

Compute clusters and HPC workloads

This card suits operators who need GPU-accelerated compute in supported HP server platforms. It fits clusters managed with HP Cluster Management Utility for driver provisioning and health monitoring. Buyers who require professional visualisation alongside compute should evaluate the M2070Q variant instead. Customers without compatible ProLiant hardware or CUDA-aware scheduling software should look at other accelerator types.

6 GB GDDR5 with ECC

The 6 GB GDDR5 frame buffer includes error-correcting code memory to protect scientific and financial datasets. This capacity and memory type determine whether the card can hold the working set for a given CUDA application. Verify that the target workload fits within 6 GB before purchase.

Highlights

  • 6 GB GDDR5 memory with ECC for large dataset handling
  • 3 Tflop single-precision peak for accelerated compute workloads
  • 655 Gflop double-precision peak for scientific accuracy
  • CUDA programming interface enables GPU-aware cluster scheduling
  • Designed for integration with HP ProLiant servers

Specifications

BrandHPE

Questions about this item

Does the HP M2075GPU Tesla graphics card ship with any cables, brackets or software media, or must they be sourced separately?

The card is supplied as a standalone module; no cables, brackets or installation media are included in the box.

How do I decide between the M2075 and the M2070Q when both appear in the same server line?

Choose the M2075 for pure compute workloads; the M2070Q adds Quadro-class visualisation features on the same GPU.

What does the quoted 1.3 Tflop single-precision figure mean for actual application performance?

It represents peak theoretical throughput; real-world results depend on memory bandwidth, kernel efficiency and the specific CUDA workload.

Also in this aisle

People compared these

Same shelf, same checkout — eight coins and a 30-minute rate lock.