SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2401.09670 · 2024

DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 3 failures of our instrument; 0 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2401.09670.json · badge: /api/ran/2401.09670.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
LLMServe/DistServe canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
FunctionStatusWhere it lives
init_logger Ran
ran; our contract draft was wrong, not the code
LLMServe/DistServe/distserve/logger.py
code served (permissive licence) · get_code("197a0f338024eca9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned Logger. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import logging

def init_logger(name: str):
    return logging.getLogger(name)

ARGS = {
    "name": "test_logger"
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Logger — a Logger output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Logger output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
json_decode_lifetime_events Ran
ran; our contract draft was wrong, not the code
LLMServe/DistServe/distserve/lifetime.py
code served (permissive licence) · get_code("465ffc17fb0aa64d")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import json
import time
from enum import Enum
from typing import List

class LifetimeEventType(Enum):
    """
    The type of an event in a request's lifetime
    """
    Issued = 'issued'
    ContextBegin = 'context_begin'
    ContextEnd = 'context_end'
    MigrationBegin = 'migration_begin'
    MigrationEnd = 'migration_end'
    DecodingBegin = 'decoding_begin'
    DecodingEnd = 'decoding_end'

    def __str__(self) -> str:
        return self.value

class LifetimeEvent(json.JSONEncoder):
    """
    An event in a request's lifetime
    Contains a timestamp and a type
    """

    def __init__(self, event_type: LifetimeEventType, timestamp: float=None):
        if timestamp is None:
            timestamp = time.time()
        self.event_type = event_type
        self.timestamp = timestamp

def json_decode_lifetime_events(json_list: List[dict]) -> List[LifetimeEvent]:
    return [LifetimeEvent(LifetimeEventType(event['event_type']), event['timestamp']) for event in json_list]

# Setup code to construct valid arguments
json_list = [
    {
        'event_type': 'issued',
        'timestamp': 1000.5
    },
    {
        'event_type': 'context_begin',
        'timestamp': 1001.2
    },
    {
        'event_type': 'decoding_end',
        'timestamp': 1002.8
    }
]

ARGS = {
    'json_list': json_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
json_encode_lifetime_events Ran
ran; our contract draft was wrong, not the code
LLMServe/DistServe/distserve/lifetime.py
code served (permissive licence) · get_code("0ce83e175d41588b")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 1, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import numpy as np
from enum import Enum
import time
import json
from typing import List

class LifetimeEventType(Enum):
    """
    The type of an event in a request's lifetime
    """
    Issued = 'issued'
    ContextBegin = 'context_begin'
    ContextEnd = 'context_end'
    MigrationBegin = 'migration_begin'
    MigrationEnd = 'migration_end'
    DecodingBegin = 'decoding_begin'
    DecodingEnd = 'decoding_end'

    def __str__(self) -> str:
        return self.value

class LifetimeEvent(json.JSONEncoder):
    """
    An event in a request's lifetime
    Contains a timestamp and a type
    """

    def __init__(self, event_type: LifetimeEventType, timestamp: float=None):
        if timestamp is None:
            timestamp = time.time()
        self.event_type = event_type
        self.timestamp = timestamp

def json_encode_lifetime_events(events: List[LifetimeEvent]) -> List[dict]:
    return [{'timestamp': event.timestamp, 'event_type': str(event.event_type)} for event in events]

# Create sample LifetimeEvent objects
event1 = LifetimeEvent(LifetimeEventType.Issued, timestamp=1000.5)
event2 = LifetimeEvent(LifetimeEventType.ContextBegin, timestamp=1001.2)
event3 = LifetimeEvent(LifetimeEventType.DecodingEnd, timestamp=1002.8)

events_list = [event1, event2, event3]

ARGS = {
    'events': events_list
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 1, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 2 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
init_logger 5 other papers
5 later
2403.02310 (2024-03)
Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Serve
2404.02015 (2024-04)
MuxServe: Flexible Spatial-Temporal Multiplexing for Multiple LLM Serving
2405.05465 (2024-05)
Vidur: A Large-Scale Simulation Framework For LLM Inference
and 2 more not named here

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

DistServe improves the performance of large language models (LLMs) serving by disaggregating the prefill and decoding computation. Existing LLM serving systems colocate the two phases and batch the computation of prefill and decoding across all users and requests. We find that this strategy not only leads to strong prefill-decoding interferences but also couples the resource allocation and parallelism plans for both phases. LLM applications often emphasize individual latency for each phase: time to first token (TTFT) for the prefill phase and time per output token (TPOT) of each request for the decoding phase. In the presence of stringent latency requirements, existing systems have to prioritize one latency over the other, or over-provision compute resources to meet both. DistServe assigns prefill and decoding computation to different GPUs, hence eliminating prefill-decoding interferences. Given the application's TTFT and TPOT requirements, DistServe co-optimizes the resource allocation and parallelism strategy tailored for each phase. DistServe also places the two phases according to the serving cluster's bandwidth to minimize the communication caused by disaggregation. As a result, DistServe significantly improves LLM serving performance in terms of the maximum rate that can be served within both TTFT and TPOT constraints on each GPU. Our evaluations show that on various popular LLMs, applications, and latency requirements, DistServe can serve 7.4x more requests or 12.6x tighter SLO, compared to state-of-the-art systems, while staying within latency constraints for > 90% of requests.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2401.09670")
get_code_for_paper("2401.09670")
have("2401.09670")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2401.09670.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2401.09670.svg)](https://syntology.ai/paper/2401.09670)

Connect an agent — have() is free.