I am working on Nifi flows for my team and facing issues on managing the performance of flow while using SplitText/ReplaceText/ExtractText.
One of the workaround i got is to use 3-4 SplitText Processors, but not sure how can we do the same.
Please let me know what changes i required in my flows to ingest and transform 1 GB of data records.
PS. Please refer this link to know all of my use cases -
Thanks in advance!
Could you post more about the following:
Here are some recommended hardware configs based on throughput & etc. can you verify you adhere (or at least close) to the reocmmendations?
|Number of NiFi nodes||CPU Cores/node||Number of disks/node, size of each disk (RAID 5/10)||RAM/node||Ideal Networking Setup|
|3||16+||6+, 1TB||8+ GB||1 Gigabit bonded NICs|
|5||16+||6+, 2TB||8+ GB||1 Gigabit bonded NICs|
|7||24+||12+, 4TB||16+ GB||10 Gigabit bonded NICs|
|9||24+||12+, 8TB||16+ GB||10 Gigabit bonded NICs|