4. Performance Metrics and Perspectives Flashcards
1
Q
System-Centric Performance Metrics
A
- Evaluate how well the overall system or any subset of its resources are being used, regardless of the quality of service
2
Q
Process-Centric Performance Metrics
A
- Evaluate how well a particular resource consumer is being served, regardless of the overall system resource efficiency
3
Q
Utilization
A
- The percentage of time that resource S was in use over time T
- Resource specific metric
4
Q
Throughput
A
- The number of requests that are processed by the system per unit time
5
Q
Effective Throughput (Goodput) (Bandwidth)
A
- The number of requests of value that are completed over a period T
6
Q
System Capacity
A
- The maximum throughput that a system can sustain
7
Q
Bottleneck
A
- The first resource to reach 100% utilization
8
Q
Response Time (Latency)
A
- The time elapsed from when the request is sent until the service is finished
9
Q
Lag
A
- A measure of how much later after the deadline service was completed, on average
10
Q
Average Lag
A
- The mean lag over all tasks
11
Q
Average Lateness
A
- The mean lag for late tasks
12
Q
Starvation
A
- When a process is not serviced at all while others are continuously serviced
13
Q
Fairness
A
- For tasks with identical priority, a fair system will try to keep the ratio of response time and service time close to each other
14
Q
Speedup
A
T_old / T_new
15
Q
Slowdown
A
T_new / T_old
16
Q
Amdahl’s Law (Equation)
A
- If a resource is used while serving requests for a fraction of time f, making the resource x-times faster will improve overall performance by a factor of y
- y = 1 / (1 - f * (1 - 1/x))
17
Q
Amdahl’s Law (Qualitative)
A
- Speeding up the bottlenecked resource will result in the most improvement
18
Q
Amdahl’s Law (Parallelism)
A
- If 1/N of the total time taken by a program cannot be parallelized, the best you can achieve is to execute the program in 1/N of the original time
19
Q
Speedup (Parallelization)
A
- The ratio of the time it takes to execute the program with 1 processor and the time it takes to execute the program with j processors
- T(1) / T(j)
20
Q
Reliability
A
- The probability that the system functions correctly continuously for a specified period
- R(t) = e^(-t/MTBF)
21
Q
Availability
A
- The probability that a system is available (“up”) at an arbitrary point in time
- A = MTBF / (MTBF + MTTR)
22
Q
Mean Time Between Failures (MTBF)
A
- The average amount of time between a completed repair and the next failure
- MTBF = T_up / n_fail
23
Q
Mean Time To Repair (MTTR)
A
- The average amount of time for a failed system to be repaired
- MTTR = T_down / n_fail
24
Q
Predictability
A
- The difference between best/average-case response times and the worst-case response time