Papers › SpotServe: Serving Generative Large Language Models on Preemptible Instances

SpotServe: Serving Generative Large Language Models on Preemptible Instances

27 Nov 2023arXiv:2311.15566archive 2025-07-28

Xupeng Miao, Chunan Shi, Jiangfei Duan, Xiaoli Xi, Dahua Lin, Bin Cui, Zhihao Jia

The high computational and memory requirements of generative large language models (LLMs) make it challenging to serve them cheaply. This paper aims to reduce the monetary cost for serving LLMs by leveraging preemptible GPU instances on modern clouds, which offer accesses to spare GPUs at a much cheaper price than regular instances but may be preempted by the cloud at any time. Serving LLMs on preemptible instances requires addressing challenges induced by frequent instance preemptions and the necessity of migrating instances to handle these preemptions. This paper presents SpotServe, the first distributed LLM serving system on preemptible instances. Several key techniques in SpotServe realize fast and reliable serving of generative LLMs on cheap preemptible instances. First, SpotServe dynamically adapts the LLM parallelization configuration for dynamic instance availability and fluctuating workload, while balancing the trade-off among the overall throughput, inference latency and monetary costs. Second, to minimize the cost of migrating instances for dynamic reparallelization, the task of migrating instances is formulated as a bipartite graph matching problem, which uses the Kuhn-Munkres algorithm to identify an optimal migration plan that minimizes communications. Finally, to take advantage of the grace period offered by modern clouds, we introduce stateful inference recovery, a new inference mechanism that commits inference progress at a much finer granularity and allows SpotServe to cheaply resume inference upon preemption. We evaluate on real spot instance preemption traces and various popular LLMs and show that SpotServe can reduce the P99 tail latency by 2.4 - 9.1x compared with the best existing LLM serving systems. We also show that SpotServe can leverage the price advantage of preemptive instances, saving 54% monetary cost compared with only using on-demand instances.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 10 code samples harvested from 1 repository linked to this paper: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code; here there are none.

By repository: official repository: 10 samples from 1 repository, 10 ran (10 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

hsword/spotserve officialmentioned in papermentioned on GitHubApache-2.0Syntology: 10 samples, 10 ran (0 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

10 samples harvested: 10 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 0 honoured, 0 violated, 10 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

10ran

Licence: 0 of the 10 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from hsword/spotserve. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

hsword/spotserve official repository 10 samples, 10 ran (0 instrument-failed) · 0 unverified
ceil_div in 5 papers elastic-switch/global_server/HotSwitch.py ran fingerprinted Apache-2.0 (permissive) · 30d6d8757af191c7 · report fixture: signature a:int/0/int, b:int/0/int · returned int · level recorded 2026-09-24
get_coord_map elastic-switch/scheduler/commands.py ran Apache-2.0 (permissive) · 469ef81be369994b · report fixture: signature hosts:list/1/Host · returned dict · level recorded 2026-09-24
get_graph_weights elastic-switch/global_server/KMmatcher.py ran Apache-2.0 (permissive) · 3554defde27c78c0 · report fixture: signature old_stages:list/1/int, old_s:list/1/int, new_stages:list/1/int, new_s:list/1/int, model_param_config:tuple/1/int, buff_config:tuple/1/int · returned float · level recorded 2026-09-24
get_hosts_info elastic-switch/scheduler/commands.py ran Apache-2.0 (permissive) · d1d23eda2dde7d6d · report fixture: signature hosts:list/1/VNode · returned tuple · level recorded 2026-09-24
get_param_client_cmd elastic-switch/scheduler/commands.py ran Apache-2.0 (permissive) · 36b63ed0075db3f2 · report fixture: signature hosts:list/1/VirtualNode, cfg:str/0/str, ckpt_path:str/0/str, strategy:tuple/1/str · returned str · level recorded 2026-09-24
get_rsync_FT_cmd sync_code.py ran fingerprinted Apache-2.0 (permissive) · a9922cd0f0c03fd8 · report fixture: signature ip:str/0/str · returned str · level recorded 2026-09-24
get_rsync_GS_cmd sync_code.py ran fingerprinted Apache-2.0 (permissive) · 27ab0edc3f0a44bd · report fixture: signature ip:str/0/str · returned str · level recorded 2026-09-24
get_rsync_PC_cmd sync_code.py ran fingerprinted Apache-2.0 (permissive) · eb3e7526ed0655ae · report fixture: signature ip:str/0/str · returned str · level recorded 2026-09-24
layer_decode elastic-switch/global_server/HotSwitch.py ran fingerprinted Apache-2.0 (permissive) · 16af849f943a512b · report fixture: signature layer_encode_id:int/0/int · returned tuple · level recorded 2026-09-24
layer_encode elastic-switch/global_server/HotSwitch.py ran fingerprinted Apache-2.0 (permissive) · e366e951d9c9c3a4 · report fixture: signature layer_id:int/0/int, tp_stage:int/0/int, tp_grain:int/0/int · returned int · level recorded 2026-09-24

Tasks

Graph Matching

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections