For the first question I would like to give one simple example
of a supercar:
sustained rate was used:
- A supercar can have insane top
speed but one can't maintain it for a long time because of the
overhead on the engine, One can also understand this by Usain
Bolt's example.
- His speed can be only helpful in
100m race, in the marathon, he might not able to manage his top
speed at all times.
- The same is applicable to HDD/SDD,
in all most all scenario sustained rate is used, you can check
HDD/SDD read/write usage in the Task Manager performance tab (if
using windows), even if HDD is utilizing 100% Disk usage at a given
time(like if you are using too many Chrome Tabs), it is unlikely to
hold this rate because it deteriorates the quality of storage
device due to more overhead. 1
- So in general while doing simple
work like watching any videos or working on any document, most
probably sustained rate is used.
burst rate was
"triggered":
- Burst Rate is used when there is a
huge amount of important data, which can not be stored on temporary
storage devices like RAM, the burst rate is triggered to transfer
this data to persistent storage burst buffer space.
- This data is transferred too
quickly to Burst buffer and but Burst buffer is not the data's
destination.
- After receiving this data from
temporary storage it transfers this data to Persistent storage like
HDD at normal speed.
- The problem with Temporary storage
is it can be erased due to any abnormality so it is required to
transfer this important data to permanent storage as soon as
possible, this is when the Burst rate is triggered, allocating
continuous storage big chunks of data is transferred
parallelly.
- For example, when a system
interrupts occurs and the processor is doing critical computation,
at this time interrupt is a priority but data is also needed to be
stored somewhere for further processing, at times like this burst
rate is triggered.
Thank you.