Node start process very depends from amount of data in blobovniczas #698
Labels
No labels
P0
P1
P2
P3
badger
frostfs-adm
frostfs-cli
frostfs-ir
frostfs-lens
frostfs-node
good first issue
triage
Infrastructure
blocked
bug
config
discussion
documentation
duplicate
enhancement
go
help wanted
internal
invalid
kludge
observability
perfomance
question
refactoring
wontfix
No project
No assignees
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: TrueCloudLab/frostfs-node#698
Loading…
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Expected Behavior
We need to make node start time as least as possible.
Current Behavior
If system filled by 7% node start process takes 17min.
If system filled by 10% node start process takes 35min.
The following step (opening blobovnicaz) takes the longest time:
At the moment if we fill system to 80-90% start node takes a huge time.
Version
Your Environment
HW
Node start process very depends from size of data in blobovnicazto Node start process very depends from amount of data in blobovnicazWe can parallelize this on a blobovnicza level (already parallelized on shard level).
Also, let's check how much does it take to open a very huge db. It may be beneficial first opening it in readonly (See
1105f7573e/db.go (L80)
, but this needs investigation).Node start process very depends from amount of data in blobovnicazto Node start process very depends from amount of data in blobovniczas