[New blog] Inside vLLM: Anatomy of a High-Throughput LLM Inference System #80
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
Deploying vllm-blog-source with
|
| Latest commit: |
3476bef
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://0b9d3992.vllm-blog-source.pages.dev |
| Branch Preview URL: | https://gordicaleksa-anatomy-vllm.vllm-blog-source.pages.dev |
|
Ok i see i haven't done the DCO thing, let me fix that. |
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
gordicaleksa
force-pushed
the
gordicaleksa/anatomy-vllm
branch
from
76ff8b4 to
c16a69e
Compare
gordicaleksa
changed the title
|
Very nice!! Any way we can get footnote working properly? |
|
@simon-mo which footnote? not sure i understood? |
|
Ah sorry I meant citations [1] [2]... they don't link the actual references. |
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
gordicaleksa
force-pushed
the
gordicaleksa/anatomy-vllm
branch
from
293aea1 to
66b6d03
Compare
|
oh ok, added! |
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
Signed-off-by: Aleksa Gordic <gordicaleksa@gmail.com>
|
@simon-mo lmk what you think now and happy to merge! |
youkaichao
reviewed
Sep 6, 2025
| <ol type="a"> | ||
| <li>policy setting - it can be either <b>FCFS</b> (first come first served) or <b>priority</b> (higher priority requests are served first)</li> | ||
| <li><code>waiting</code> and <code>running</code> queues</li> | ||
| <li>KV cache manager - the heart of paged attention [[3]](#ref-3)</li> |
There was a problem hiding this comment.
this reference rendering seems to be broken inside <li></li>
youkaichao
reviewed
Sep 6, 2025
|
|
||
| > [!NOTE] | ||
| > Block size for a standard transformer layer (non-MLA [[4]](#ref-4)) is computed as follows: | ||
| > 2 * <code>block_size</code> (default=16) * <code>num_kv_heads</code> * <code>head_size</code> * <code>dtype_num_bytes</code> (2 for bf16) |
There was a problem hiding this comment.
the first 2 is for k and v right? 2 for bf16 is for dtype_num_bytes
youkaichao
approved these changes
Sep 6, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Porting the original blog: https://www.aleksagordic.com/blog/vllm