OptoCloud: Ultra-fast optically interconnected heterogeneous Data Centers

The majority of human activities, including transport, Internet, banking, public health and entertainment, depend on Data Centers. Cloud traffic is forecasted to grow exponentially and account for 95% of global traffic. In 2015, the total power consumption of data centers worldwide was higher than the national power consumption of the UK and is predicted to increase up to 15-times by 2030.

Currently, all data center networks are formed based on hierarchical electronic packet switched networks; however, they can't keep up with demand creating a ever increasing gap between data growth and Moore's Law. So, while compute node power, measured in flop/s, has increased by 65 times in the last 18 years, the node communication bandwidth has only increased by 4.8 times and the bytes communicated per flop have decreased 8 times. This creates a computation to communication wall, minimizing data movement and constraining applications to operate locally. In addition, these systems also suffer from very high median latencies, O(100microseconds) (order of 100microseconds), and 99.9-percentile tail latencies, O(100ms), to the detriment of the system and application performance.

The OptoCloud fellowship aims to design and build an energy efficient, cost effective, scalable, single hop, and nanosecond speed optical circuit switched network. This will interconnect heterogeneous systems made of servers, CPUs, accelerators, neuromorphic processors, memory elements, storage to support different parts (rack, end-of-row) and sizes of data centers (small-medium size ~10-100,000 to ~1,000,000 server farm). Crucially, the network aims to offer zero data loss, without in-network a) buffering, b) active switching and routing, and c) network header addressing and processing to minimize complexity, and to consume very low power. Furthermore, the system also will inherently support 1-to-1, 1-to-N, N-to-N and N-to-1 connectivity in a synchronous manner without the need for data replication for multi/broad -casting, currently not possible. This is key to support diverse workloads such as storage caching, large-scale database lookups, training distributed deep neural networks, parallel computing that use communication primitives such as allreduce, broadcast and reduce, gather and scatter, all-to-all among others.

To achieve these, OptoCloud will explore the fundamental challenges of sub-nanosecond optical switching, near receiver-less low-power transceivers and nanosecond scheduling able to reconfigure circuits and shape IT and network topologies every 10s-100s of nanoseconds. It aims to offer orders of magnitude improvement in a) switching, b) scheduling and network topology re-configuration, c) power consumption, d) medium and tail latency and finally e) throughput with zero data loss.

The PI will work with the PDRAs, PhD students, industrial partners (Microsoft, Finisar, Xilinx, Sumitomo Electric), as well as universities (Columbia and National Technical University of Athens) and form a unique compute and optical network ecosystem to methodologically answer fundamental questions while reflecting all necessary requirements on the proposed concepts, and rigorously evaluating developed technologies using industrial driven use case scenarios.

The technologies proposed will provide the means for the design and implementation of a new form of computer and network architecture. This is the heterogeneous and disaggregated Data Center system where the network technologies proposed at its core will unlock its potential to deliver unparalleled modularity and performance. The proposed technologies can support the increasing data volume, diversity and unpredictability of connected computing systems while reducing power consumption and CO2 footprint. It will enable accelerated creation and innovation of new services and applications as well as solve scientific problems currently not possible due to the rigid data centre and high performance computer architecture. This will benefit everyone who uses and relies on networked technologies.

All major communication and computing stakeholders will benefit from the fellowship results.

*Technology manufacturers and vendors: The results of the fellowship will be invaluable in designing the equipment of the future to maximize performance, flexibility and programmability by benefiting from the fusion of photonic and electronic systems. The project partners Finisar, Xilinx, and Sumitomo Electric will be the most immediate beneficiaries but others will benefit.

*Data Center and High Performance Computing operators: Using the fellowship, heterogeneous reconfigurable data center architectures can be reconfigured millions of times per second to deliver maximum utilization, best serve diverse workloads and unlock the ability to perform distributed parallel computation and distributed deep neural network tasks at scale. Microsoft, a partner of this fellowship, will be a direct beneficiary, and others will follow.

*Creation of SMEs: The fellowship will stimulate the generation of future business opportunities by creating a new sector of disaggregated and heterogeneous computer and network architecture technologies. This will allow the creation of a range of SMEs that can create revenues either through the development and licensing of software/hardware function modules or delivering complete hardware solutions. The resulting business opportunities will contribute to job creation and economic prosperity.

*Impact beyond ICT: The principles, concepts and techniques developed are directly transferable to other sectors within and beyond ICT that use and benefit from highly modular computing systems. This includes but not limited to 5G and beyond networks, Internet of Things, smart cities, satellite, High Performance Computing, consumer electronics, embedded and distributed systems, robotics, fundamental engineering, manufacturing, energy, automotive, consumer electronics and health.

*Academic and research community: The fellowship can play a key role towards the realization of a new philosophy of using optical and scheduling technologies to pool together functional modules to form complete computing and network systems. This inherently creates a new research field that will stimulate fundamental rethinking on the design and operation of systems and networks as well as the creation of new programming models and application design. Columbia University and NTUA are partners, so involved directly, but the wider community will also benefit through widespread dissemination.
In collaboration with UCL's Business and industrial partners I will take advantage of existing expertise for the communication, protection and exploitation of the results.