The aim of OPERA is to create a cooperative, secure, reliable, customized, and low power computing platform that is able to address the challenges imposed by the future convergence of datacentre computing, embedded devices and sensors. To this end OPERA will develop a new generation of high-density servers. These servers are the basic “bricks” for implementing a scalable Low Power datacentre. At the base of these modules there are heterogeneous architectures that use ARM, Intel, POWER8 processors to enable energy efficient server-class processors, and FPGA accelerators for optimized functions and computation offloading (i.e., by instantiating on the reconfigurable device customized circuits that perform specific operation directly in hardware). These devices form a mix of processing elements and accelerators designed for achieving significantly better energy efficiency at the cost of flexibility. Moreover, the integration of the CAPI (Coherent Accelerator Processor Interface) technology – it allows external component such as FPGA-based circuit to have a coherent access to the cache hierarchy of a processors, allows these accelerators appearing to external high-performance systems as if they were cores integrated on the same chip. The OPERA project aims also at exploiting high-speed optical links to provide interconnections between the accelerators and the external system with no performance limitations.
These technologies forms the substrate used by OPERA to implement an efficient workload decomposition system that automatically and dynamically executes tasks on the most suitable processing element. The result of the integration of this workload decomposition system is to make optimal use of computing resources and application in a heterogeneous architecture so that it is possible to guarantee high performance computing capabilities in a scalable (cloud) energy efficient infrastructure perspective.
In order to exploit at all levels improvements in computer systems made of advanced components, the project will develop a highly parallel embedded system based on an ultra-low power System-on-Chip (SoC) which can be deployed on various application domains. The strong integration of sensors, radio communication interfaces, and computing resources allow spreading such a small intelligent devices in several contexts. The availability of reconfigurable communication interfaces (e.g., wireless vs. wired) guarantees that these devices can be flexibly adapted to different contexts (e.g., smart cameras can dynamically switch between Wi-Fi and 3G connections depending on the quality of the signal with the aim of saving energy every time it is necessary). Highly parallel on-board processing elements provide enough computing power for pre-processing raw data that can be later sent to a remote facility (scalable Low Power datacentre) for subsequent analysis. OPERA leverages these capabilities to deliver real-world applications where to test the proposed platform and that can easily be extended to other application scenarios.
The breakthrough proposed by the OPERA project is to ensure an engineering research innovation in the field of advanced computing through:
(i) a structural change on the computing workload decomposition,
(ii) the integration of specialized advanced hardware components managed by software aiming at increasing the efficiency of high performance heterogeneous systems, and
(iii) the implementation of metrics for the evaluation and subsequent management of energy efficiency in integrated complex computing systems.
In order to ensure to meet the key elements of the proposed computing platform, the project will be driven by industrial, technology, and scientific requirements in vertical domains: ultra-low power architectures, innovative power management, highly-parallel computer systems, energy efficiency, and software components. All of these vertical domains will ensure the low-power, low-cost, high density, secure and reliable aspects that are stringent requirements in the H2020 Work Programme.
OPERA project propose to provide a customized and low power computing hardware software integrated solutions based on three strategic axes:
1. Architecture: heterogeneous, reconfigurable, low power and ultra-low power integrated components;
2. Computing platform: including hardware, software, and communication for a secure and reliable system;
3. Energy efficiency: software layer for monitoring energy efficiency of the computing platform components.