Move TSC took-time policy to guard both heap and disk tier #17190
Conversation
Signed-off-by: Peter Alfonsi <[email protected]>
peteralfonsi
requested review from
anasalkouz,
andrross,
ashking94,
Bukhtawar,
CEHENKLE,
cwperks,
dblock,
dbwiddis,
gbbafna,
jainankitk,
kotwanikunal,
linuxpi,
mch2,
msfroh,
nknize,
owaiskazi19,
reta,
Rishikesh1159,
sachinpkale,
saratvemulapalli,
shwetathareja,
sohami and
VachaShah
as code owners
github-actions
bot
added
enhancement
Signed-off-by: Peter Alfonsi <[email protected]>
peteralfonsi
added
backport
|
❌ Gradle check result for 36c7ba2: FAILURE Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change? |
Signed-off-by: Peter Alfonsi <[email protected]>
|
❌ Gradle check result for 0359de4: FAILURE Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change? |
| */ | ||
| private static final Setting.AffixSetting<TimeValue> TIERED_SPILLOVER_DISK_TOOK_TIME_THRESHOLD = Setting.suffixKeySetting( | ||
| private static final Setting.AffixSetting<TimeValue> TIERED_SPILLOVER_TOOK_TIME_THRESHOLD = Setting.suffixKeySetting( |
There was a problem hiding this comment.
I think it would be better to introduce a new setting for overall Tiered cache rather than reusing existing one. As it would be very confusing to users/anyone seeing the setting name which otherwise works on both onHeap/disk tier.
We could eventually decide to deprecate this(disk policy setting) altogether later if needed.
Signed-off-by: Peter Alfonsi <[email protected]>
Signed-off-by: Peter Alfonsi <[email protected]>
|
❌ Gradle check result for 294ebae: FAILURE Please examine the workflow log, locate, and copy-paste the failure(s) below, then iterate to green. Is the failure a flaky test unrelated to your change? |
Description
Changes the TieredSpilloverCache's minimum took-time policy so that queries must take >10 ms to enter either tier, instead of just to enter the disk tier. This is desirable because when we allow caching size > 0 queries in the request cache, the number of cacheable queries may go way up, and we should avoid flooding the heap tier with overly cheap queries.
Note the setting key
tiered_spillover.disk.store.policies.took_time.thresholdis unchanged, for backwards compatibility.Related Issues
Resolves #16162
Check List
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.
For more information on following Developer Certificate of Origin and signing off your commits, please check here.