Sleep mode #252
Merged
Sleep mode #252
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: irar2 <[email protected]>
Signed-off-by: irar2 <[email protected]>
Signed-off-by: irar2 <[email protected]>
Signed-off-by: irar2 <[email protected]>
mayabar
requested changes
Nov 6, 2025
pkg/kv-cache/block_cache.go
Outdated
| } | ||
|
|
||
| func (bc *blockCache) start(ctx context.Context) { | ||
| bc.logger.V(logging.TRACE).Info("Starting KV cache") |
Collaborator
There was a problem hiding this comment.
this log message is not supposed to be printed a lot, consider to report on INFO level
pkg/kv-cache/block_cache.go
Outdated
| } | ||
|
|
||
| func (bc *blockCache) discard() { | ||
| bc.logger.V(logging.TRACE).Info("Discarding KV cache") |
pkg/kv-cache/block_cache.go
Outdated
| } | ||
|
|
||
| func (bc *blockCache) activate() { | ||
| bc.logger.V(logging.TRACE).Info("Activating KV cache") |
| "github.com/vmihailenco/msgpack/v5" | ||
| ) | ||
|
|
||
| func ParseKVEvent(parts [][]byte, expectedTopic string, expectedSeq uint64) ([]uint64, []uint64, bool) { |
Collaborator
There was a problem hiding this comment.
maybe move this function to common.test_utils like CreateSub?
| } else if string(tags) == "kv_cache" { | ||
| wakeUpKVCache = true | ||
| } | ||
| } |
Collaborator
There was a problem hiding this comment.
in case tags are not defined, both wakeUpWeights and wakeUpKVCache should be true
pkg/llm-d-inference-sim/server.go
Outdated
| // Activate the kv cache if either the sim is sleeping and the wake up tags are not | ||
| // "only weights", or the sim is not sleeping but the cache is still disabled | ||
| // and the tags are "only kv cache" | ||
| if s.config.EnableKVCache && (s.isSleeping && !wakeUpWeights) || (!s.isSleeping && s.kvCacheDisabled && wakeUpKVCache) { |
Collaborator
There was a problem hiding this comment.
shouldn't it be s.config.EnableKVCache && (....)?
pkg/llm-d-inference-sim/simulator.go
Outdated
| // variable VLLM_SERVER_DEV_MODE | ||
| isInDevMode bool | ||
| // in sleep mode, and if woken up with weights only, kv cache is disabled | ||
| kvCacheDisabled bool |
Collaborator
There was a problem hiding this comment.
this is a duplication of bloockCach.disabled?
Signed-off-by: irar2 <[email protected]>
Signed-off-by: irar2 <[email protected]>
mayabar
approved these changes
Nov 6, 2025
Collaborator
|
/lgtm |
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Closes #218