System Architecture and Reliability

In a semiconductor fab, the aim is to detect abnormalities quickly using sensor data collected at high frequency from many pieces of manufacturing equipment. Instead of sending all the data to a distant cloud for processing, edge computing processes it on computers placed near the equipment. Which is a main advantage of using edge computing?

  1. AManagement of the computers placed near the equipment becomes unnecessary
  2. BCommunication delay is small, so the system can respond immediately, and the amount of data sent far over the network can also be reduced
  3. CProcessing capacity can be increased without limit compared with the cloud
Show answer and explanation

Answer: B. Communication delay is small, so the system can respond immediately, and the amount of data sent far over the network can also be reduced

Edge computing processes data close to where it is generated. Because it does not wait for round trips to a distant cloud, it can respond immediately for abnormality detection and equipment control, and network traffic can be reduced by sending only the needed results or summaries. In exchange, the computers placed at each site must be managed and their processing capacity cannot be increased as freely as in the cloud, so it is generally combined with the cloud for tasks such as long-term data analysis.

  • A. Managing the computers placed at each site actually becomes necessary.
  • B. Correct. Shorter delay and less data to send are the main advantages.
  • C. It is rather the cloud that makes it easy to increase processing capacity flexibly.
Practice this topic Back to topic

Related calculators

Books on System Architecture and Reliability Ad

As an Amazon Associate I earn from qualifying purchases.

More System Architecture and Reliability questions

  1. Among the ways of connecting storage, which describes a SAN (storage area network)?
  2. From a storage pool with a physical capacity of 100 TB, 30 volumes with a logical capacity of 5 TB each were created using thin provisioning. Currently, on average 40% of each volume's logical capacity has been written, and the data grows by a total of 2 TB each week. Ignoring data deletion and compression, when will the pool's utilization reach 80%?
  3. A data center has IT equipment consuming 800 kW, cooling facilities consuming 280 kW, and power facility losses, lighting and so on amounting to 120 kW. When the cooling method is improved so that only the power consumption of the cooling facilities is halved, which is the PUE?
  4. Which is the reason for combining a UPS (uninterruptible power supply) with on-site generators in a data center?
  5. Which describes the hot aisle/cold aisle concept used in arranging server racks in a data center?