Continued from the previous post.

With the graphs below you can compare calculation time and cost for a fixed amount of calculations in Floating Point Operations (FLOPs^{***}). Use buttons above the graphs to set calculations amount and number of virtual or bare metal machines called “nodes” used for calculations.

Important notice: we assume that a task can be run on multiple computers WITHOUT any slowdown. This means that on N machines the task will finish N times faster. This could be true, for instance, in case of hyperparameters search when you have multiple independant tasks.

Graphs are not scaled when parameters change to make points movements clearly visible. To scale graphs manually use “Autoscale” and “Reset axes” buttons that appear in the top right corner of the graphs when you bring the mouse cursor over it (on tablet devices tap the graph).

### Filter offers in all graphs below using these charts

### Set calculation task complexity in EFLOPs and nodes number

The graph below shows calculation time and cost necessary for calculating a task with task complexity set in EFLOPs^{****} on GPU(s) provided by each offer.

### Set calculation task complexity in EFLOPs and nodes number

The graph below shows calculation time and cost necessary for calculating a task with the same complexity using only provided CPUs.

* All sums are in USD. Offers based on other currencies are converted using today’s rates from European Central Bank:

All prices do not include taxes.

** Performance data shows single precision FLOPS for GPUs and single precision FLOPS for CPUs. **Beware that performance data is a rough estimate, and actual performance may vary greatly between applications.**

*** Floating Point Operations (not to be confused with Flops — Floating Point Operations per Second).

**** EFLOPs = 1 * 10^{18}FLOPs. TFLOPs = 1 * 10^{12}FLOPs.

## One thought on “Cloud GPU providers comparison – more graphs”