M2DC Image appliance

We optimise your costs in photo finishing business

M2DC IMAGE at a Glance

This M2DC appliance enables competitive advantage with respect to performance, scalability, efficiency and cost-savings to applications in need of image scaling processes. This task has become a typical use case in modern web applications like social networks, cloud applications and web shop systems, delivering a large amount of images in different sizes through responsive websites and several clients. Our success stories include Europe's largest photo finisher - CEWE - handling millions of images per year for delivering high quality photo products.

491%

of energy efficiency gain could be achieved compared to the previously used standard systems

73%

gains in the system upgrade and significant reduction of operation costs than comparable standard solutions

59%

of improved cost of ownership after five years including operational expenses

Have a look at the demonstrator!

Want to learn more?
Try it out!

GitLab repository: http://gitlab.m2dc.eu/WP6/img-appliance

We improve image scaling for you

Present-day servers are equipped with state-of-the-art general-purpose CPUs with x86 architecture, designed for high performance at workload peaks. Aiming improvement of processing performance, the M2DC IMG appliance offers an alternative based on specialized hardware, in terms of graphical processing units (GPUs). For image processing use cases they offer more performance or lower power consumption compared to standard CPU architectures. As this approach requires a specialized customization effort to make the most of the hardware, the IMG appliance couples an energy efficient implementation with an abstracted and easy to use REST interface.

Available built-in efficiency enhancements

Find a detailed list of built-in enhancements with the descriptions here.

The appliance includes a central component for increasing the energy efficiency of the image processing system

It includes a data mining enhancement to minimize the continuous workload forecastings that enables aggressive application scheduling while still meeting service level agreements

Usage of CPUs to increase energy efficiency

Usage of GPUs to increase energy efficiency by a factor of 8 compared to x86, with an additional factor using the intelligent workload management extension of the OpenStack cloud management

Appliance configuration

You can choose from a wide range of modules, which together form a heterogeneous cluster. These may be microservers with high-performance x86 or ARM64 as well as low-power ARM processors with the latter also providing GPU computing capabilities. And, in addition to the CPU-based microservers, also FPGAs are available for specialized applications. PCIe extension cards are also supported for even further flexibility.