PNY TCSK20M-PB Leaflet

TESLA K20M GPU COMPUTING ACCELERATOR 

Designed for double precision applications and the broader supercomputing market, the Tesla K20 delivers 3x the double precision performance 

compared to the previous generation Fermi-based Tesla M2090, in the same power envelope. Tesla K20 features a single GK110 Kepler GPU

that includes the Dynamic Parallelism and Hyper-Q features. With more than one teraflop peak double precision performance, the Tesla K20 

is ideal for a wide range of high performance computing workloads including climate and weather modeling, CFD, CAE, computational physics, 

biochemistry simulations, and computational finance.

With teraflops of single and double precision performance, NVIDIA® Kepler GPU Computing Accelerators are the world’s fastest and most 

efficient high performance computing (HPC) companion processors. Based on the Kepler compute architecture, which is 3 times higher per-

formance per watt than the previous “Fermi” compute architecture

1

, the Tesla Kepler GPU Computing Accelerators make hybrid computing 

dramatically easier, and applicable to a broader set of computing applications. NVIDIA Tesla GPUs deliver the best performance and power 

efficiency for seismic processing, biochemistry simulations, weather and climate modeling, image, video and signal processing, computational 

finance, computational physics, CAE, CFD, and data analytics. 

The innovative design of the Kepler compute architecture includes:

>> SMX (streaming multiprocessor) design that delivers up to 3x more performance per watt compared to the SM in Fermi. It also delivers 1 

petaflop of computing in just 10 server racks.

>> Dynamic Parallelism capability that enables GPU threads to automatically spawn new threads. By adapting to the data without going back 

to the CPU, it greatly simplifies parallel programming and enables GPU acceleration of a broader set of popular algorithms, like adaptive mesh 

refinement (AMR), fast multipole method (FMM), and multigrid methods.

>> Hyper-Q feature that enables multiple CPU cores to simultaneously utilize the CUDA cores on a single Kepler GPU, dramatically increasing 

GPU utilization, slashing CPU idle times, and advancing programmability. Ideal for cluster applications that use MPI.

TESLA K20M for Servers - PRODUCT SPECIFICATIONS

CUDA PARALLEL PROCESSING CORES

2496

PEAK DP IN TFLOPS

> 1 TFlop

INTERFACE

320-bit

PROCESSOR CORE CLOCK

706 MHz

FRAME BUFFER MEMORY

5 GB GDDR5

MEMORY BANDWIDTH

208 GB/s

DISPLAY CONNECTORS

None

MAX POWER CONSUMPTION

225 W

POWER CONNECTORS

(2) x 6-pin PCI Express power connectors

GRAPHICS BUS

PCI Express 2.0 x16

FORM FACTOR

110 mm (H) x 265 mm (L)

Dual Slot

THERMAL SOLUTION

Active

TCSK20M-PB

GPU COMPUTING ACCELERATOR

1 - Based on DGEMM performance: Tesla M2090 (Fermi) = 410 gigaflops, Tesla K20 (expected) > 1000 gigaflops

More information: www.pny.eu/tesla

PNY Technologies Europe

sales@pny.eu | T: +33 (0)5 56 13 75 75

NVIDIA

®

 TESLA

®

 K20M - Module