System Architecture and Reliability

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%?

  1. AIn 10 weeks
  2. BIn 20 weeks
  3. CIn 30 weeks
Show answer and explanation

Answer: A. In 10 weeks

The data currently written is 30 × 5 TB × 0.4 = 60 TB, which uses 60% of the 100 TB physical capacity. Only 80 − 60 = 20 TB can be added before reaching 80% (80 TB), so it takes 20 ÷ 2 = 10 weeks. Thin provisioning allocates physical capacity only for what is actually written, so here the total logical capacity of 150 TB exceeds the physical capacity. Writes become impossible once the physical capacity is used up, so monitoring utilization and adding capacity early is essential.

  • A. Correct. (80 − 60) ÷ 2 = 10 weeks.
  • B. This is the time until the pool is full (100%): (100 − 60) ÷ 2.
  • C. This is based on 80% of the total logical capacity of 150 TB (120 TB). What is actually used is the physical capacity.
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. 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?
  2. Which is the reason for combining a UPS (uninterruptible power supply) with on-site generators in a data center?
  3. Which describes the hot aisle/cold aisle concept used in arranging server racks in a data center?
  4. A web system that must process 1,800 requests per second at peak is being designed. One server can process at most 250 requests per second, and each server's utilization should be kept at 70% or less. In addition, the remaining servers must still meet this condition if one server fails (N+1 configuration). With the load spread evenly, which is the minimum number of servers required?
  5. Which describes a client-server system?