M2DC provides significant benefits for data centres, minimising their Total Cost of Ownership (TCO). This goal is achieved by the use of efficient microservers, easy and cost-effective hardware upgrade possibilities and intelligent energy management techniques integrated with OpenStack middleware.
In particular, M2DC platform enables competitive advantage with respect to performance, scalability, efficiency and cost-savings to applications such as photo finishing service processing several million images per year, data analytics of large real-time sensor data from vehicles, processing HPC workloads and providing Platform as a Service (PaaS) cloud solutions.
Detailed monitoring and intelligent management
Customised hardware
Lower costs and reduced energy consumption
Optimised appliance
The core of M2DC technology, ready to accomplish the most challenging tasks.
Demanding HPC appliance focusing cost-effectiveness and energy efficiency.
Learn moreHigh-volume, high-velocity computation of real-time analysis focusing IoT sensors and other data sources.
Learn moreEfficient and scalable image and signal processing services.
Learn moreCloud vendors with focus on Platform as a Service based on M2DC micro-server platform equipped with heterogeneous hardware.
Learn moreMeeting you where your problem is, with the most fit solution to solve it.
Images are processed to thumbnails on the fly in an energy efficient way, reducing the required data storage.
Network flows in the M2DC appliance are monitored by a high efficiency implementation of an intrusion detection system for suspicious activity to protect the appliance.
Gives M2DC appliances a very fast communication channel with very low latency that is not available neither on HP Mooshot nor bladecentres, even those using InfiniBand or similar networks.
High efficiency accelerator on FPGA microservers for convolution neural networks inference for tasks such as image classification.
High efficiency implementation of data analysis and load forecasting algorithm, to enable intelligent workload management and reduce operating costs.
High efficiency last-level scheduling cache with self-healing capabilities and customisable strategies, to efficiently map tasks to the heterogeneous resources of a M2DC appliance.
Fully programmable, fast and energy efficient pattern matching engine with a limited action domain specific language to detect problems on the stream of system monitoring events.
Symmetric and homomorphic encryption on FPGA microservers, for performance and energy efficiency of secure operations and data transferts in a M2DC appliance.
Easily program for a mix of x86 and ARM CPUs, low power and high power, in a M2DC appliance with a single system view of the microsevers memory and acceleration capabilities.
Offload collective operations to M2DC acceleration resources to reduce energy use, improve latency and optionnally decrease floating point errors.
Scalable accelerator for data mining based on self-organizing feature maps utilizing Intel Stratix 10.
Greatly efficient and with lower cost.
3 Carriers
9 High-Performance Microservers
48 Low-Power Microservers
Ready for the most demanding challenges.
9 Carriers
27 High-Performance Microservers
144 Low-Power Microservers
The most powerful option.
15 Carriers
45 High-Performance Microservers
240 Low-Power Microservers
(up to 3 Microservers)
(up to 16 Microservers)
(1 Accelerator)
Intel Xeon
32 Core Cortex-A72
Intel Stratix 10
Xilinx Zynq
Samsung Exynos
Nvidia Tegra