ARTERYS
November 1, 2022Israel
November 1, 2022Fog/Edge computing vs. Cloud computing in the Internet of things
Data handling mechanisms that help offload the cloud and make a more rapid transmission and analysis of data in wearable devices...
We live in the era of cloud computing, where most, if not all data, are stored and analyzed remotely. This data includes patients’ medical sensor data and actuator data from Artificial intelligence devices (Insulin pumps, hormonal releasers, ambulatory blood pressure monitors). This vast amount of data will undoubtedly lead to the cloud model being overwhelmed. The solution? Pushing the data processing to the network’s edge closer to data-generating devices.
Fog/ Edge computing
Fog computing and edge computing are two types of architectures for data handling. (1) They can offload data from the cloud, process it close to the patient, and rapidly transmit information machine-to-machine or machine-to-human. Sensor data is processed near the sensing and actuating devices with fog computing (with local nodes). Fog computing for medical devices allows processing data where the sensor collects data (close to the patient) rather than almost entirely up in the cloud. The edge computing paradigm moves control of a network’s services away from central nodes (defined as the core) to the other extreme, the sensor itself (defined as the edge) rather than servers or nodes. (2)
Fog computing and edge computing are two types of architectures for data handling. (1) They can offload data from the cloud, process it close to the patient, and rapidly transmit information machine-to-machine or machine-to-human. Sensor data is processed near the sensing and actuating devices with fog computing (with local nodes).
Fog computing for medical devices allows processing data where the sensor collects data (close to the patient) rather than almost entirely up in the cloud. The edge computing paradigm moves control of a network’s services away from central nodes (defined as the core) to the other extreme, the sensor itself (defined as the edge) rather than servers or nodes. (2)
Difference between Fog/Edge computing
The location of the additional computer power locus is the main difference between the two types of architecture. Fog computing assigns computing power down to a local area network (a set of interconnected computers within a limited area) where data processes within a hub, node, router, or gateway and then travel to the appropriate devices. (1) On the other hand, Edge computing assigns the processing power and communication capabilities to a data-gathering chip directly located in the device (a sensor, a detector, an embedded system, or a smart object). In some cases, data can relocate to a nearby server. Fog and edge computing can coexist and overlap in their functions within a single network of devices. Edge computing may use open-source or proprietary technologies, whereas fog almost always uses only open-source technologies.
Advantages of Fog/Edge vs. Cloud computing
Fog and edge computing will offer some advantages over the cloud:
- Greater data transmission speed.
- Less dependence on limited bandwidths.
- Greater privacy and security.
- Greater control over data generated in regulated geographical regions.
- Lower costs because of less data remotely transmitted.
Limitations of Fog/Edge computing
There are potential limitations to distributed data processing executed by fog or edge computing devices.
- Formatting data from various sources to a single typical architecture while preserving privacy if data are to be shared.
- Balancing data abstraction to facilitate limited local storage against lesser abstraction to facilitate productivity.
- Detecting unreliable data from isolated defective sensors or wireless transmitters.
Conclusion
AI wearable devices connected through the Internet have their challenges. The amount of data gathered by AI is becoming increasingly hard to handle and process, so Edge and Fog computing offer an attractive and more efficient alternative to offload the cloud. It is of utmost importance to recognize the advantages and disadvantages of these newer models and thus adopt them appropriately.