The Edge vs the Swarm
During the last decade Cloud computing has enabled the democratization of computing. Cloud computing has permitted high performant computing to be available not only in large supercomputer centers and to users who could afford to acquire expensive hardware for occasional use, it has also allowed rich computing environments to be available to all kind of businesses on-demand and in pay-per-user models, providing the illusion of cost effective, infinite resources and realizing a radical acceleration of commoditization of computing.
Computing at the Edge
Today, we witness the emergence of new forms of computing, such as Edge and Fog, which make compute resources break the barriers of the data center, and bring computing and data resources down to earth, in locations spread together with the IoT installations. These are already recognized as beneficial to avoid drawbacks of large data movements and latency. These new cloud forms provide better performance in IoT scenarios and enable wider diversity of application and services. There remains in the market significant uncertainty with regards to what is Edge computing and specifically the features and capabilities it offers. The Linux Foundation Open Edge Computing Glossary defines the Edge Cloud as “Cloud-like capabilities located at the infrastructure edge, including from the user perspective, access to elastically-allocated compute, data storage and network resources. Often operated as a seamless extension of a centralized public or private cloud, constructed from micro data centers deployed at the infrastructure edge.” Atos has recently provided its view on current status of edge computing technology, the diversity on market offerings in hardware, software, consumer and born at the edge classifications in this whitepaper.
Computing beyond Edge
Existing edge computing developments solely regard IoT devices as sources of data. Increasingly, thanks to Moore’s law, IoT endpoint devices offer more sophisticated behaviors than just to collect data carrying considerable amount of computing and storage power. Market analysists such as IDC are predicting “By 2022, over 40% of organizations' cloud deployment will include edge computing, and 25% of endpoint devices and systems will execute AI algorithms.”. For these, we foresee that edge computing is just an initial step toward even a more decentralized view of computing, Swarm computing.
Swarm computing, which we define as the new digital infrastructure encompassing rich IoT endpoints, edge and multiple cloud platforms in continuous cooperation in order to connect entities in the context of large cooperative and self-organizing applications.
A computing swarm will therefore become an opportunistic service network which consists of heterogeneous resources capable of linking dynamically, flexibly and autonomously to provide assets (i.e. services, content and resources) to relevant applications and participants. Participant resources can be IoT, edge and cloud resources. Swarms are service networks deployed and executed in response to requirements, which may emerge from application / user requests for services or information, or from opportunities offered in response to data- or event-driven activities.
From September 10th to 12th, over 400 people including Atos technical experts, partners and start-ups will gather during our 2nd Expert convention in Madrid to build the future of technology and to instil and drive technological changes that will help our clients master these challenges in a sustainable and ethical way.