Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

All 20k shall be returned

OperationTotal CM-handles registeredConcurrent requests/parallelDMI DelayRest response sizeNotesPerformanceObserved Rate
(rate not measured)

Expected duration

Sign-Off
1Registration of 100 CM-handles20k Nodes1 (requests are sequential)

100 ms to get module references
1000 ms to get module resources

N/A


1.6 ops/secondTBDTDB
2De-registration of 100 CM-handles20k Nodes1 (requests are sequential)

N/A

N/A





3CM-handle search without filter20k Nodes5N/ATBD e.g. 10MBAll 20k returnedWithin 15 sec5 ops/minute1 minute / operationTDB
4CM-handle ID search without filter20k Nodes

5

N/ATBD e.g. 1MBWithin 2 sec7.1 ops/minute42 seconds / operationTDB
5Synchronous single CM-handle pass-through read20k Nodes

20


TDB e.g. 5 KB

25 op/

sc

sec

50 ops/secondTBDTDB
6Synchronous single CM-handle response time read20k Nodes

20

N/A


within 1 sec




7Synchronous single CM-handle pass-through write20k Nodes

20

N/A

N/A

13 ops/sec 

13 ops/secondTBDTDB
8Synchronous single CM-handle response time write20k Nodes

20

N/A


Within 2 sec




...

  1. Concurrent which number are we agreeing to test, 10 parallel op. according to the input load/ , the average response time as mentioned in comment is 20 parallel op
  2. DMI delay - We need feed back from ETH on this
  3. Better to have R, W use cases separated as in the FS for both throughput and response time or should we merge this on NCMP
  4. Check Registration - Kolawole Adebisi- Refer Adeolokun  Refer to Michelle's email for the numbers
  5. Check DE-Registration if that was ever agreed

...