Operating Systems

OSDI '26

Name: OSDI '26
Start: 2026-07-13
End: 2026-07-15
Location: The Westin Seattle

The 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI '26) brings together professionals from academic and industrial backgrounds in a premier forum for discussing the design, implementation, and implications of systems software. The symposium emphasizes innovative research as well as quantified or insightful experiences in systems d…

Visit the official site

146Speakers

10Sponsors

7Sessions

70Companies · in total

Speakers

Programmed at OSDI, in alphabetical order.

Aanjhan Ranganathan

Northeastern University

Alexander Neff

Ruhr University Bochum

Amar Phanishayee

Program Co-Chair · NVIDIA

Andreas Kogler

Graz University of Technology Alumni

Andrew Chin

Georgia Institute of Technology

Antonio Bianchi

Purdue University

Antonis Louka

DistriNet, KU Leuven

Arash Ale Ebrahim

CISPA Helmholtz Center for Information Security

Atsushi Tanaka

The University of Osaka

Aurélien Francillon

EURECOM

Bertalan Borsos

Eötvös Loránd University, Hungary

Billy Brumley

Rochester Institute of Technology

Bozhen Liu

Texas A&M University - Corpus Christi

Brian Junekyu Lee

Georgia Institute of Technology

Cen Zhang

Georgia Institute of Technology

Chenguang Fang

Alibaba Group

Chengwei Li

Shanghai Jiao Tong University

Christoph Sanders

DistriNet - KU Leuven

Christos Kozyrakis

Stanford University & Nvidia Research

Connor Glosner

Purdue University

Constantinos Patsakis

University of Piraeus

Dave (Jing) Tian

Purdue University

Agenda

Selected from 7 sessions.

Monday, July 13 10:45 am–12:25 pm
Chimera: Cost-Efficient Multi-LLM Serving via GPU Memory Ballooning
KV Cache and Long Context (Track 1)
Monday, July 13 10:45 am–12:25 pm
Contextra: Hierarchical Context Caching for Long Context Language Model Serving
KV Cache and Long Context (Track 1)
Monday, July 13 10:45 am–12:25 pm
ECHO: Efficient KV Cache Offloading with Lossless Prefetching for Serving Native Sparse Attention LLMs
KV Cache and Long Context (Track 1)
Monday, July 13 10:45 am–12:25 pm
No Buffer, No Bottleneck: Efficient Zero-Copy KV Cache Offloading for Long-Context LLMs
KV Cache and Long Context (Track 1)
Monday, July 13 10:45 am–12:25 pm
Simple is Better: Multiplication May Be All You Need for LLM Request Scheduling
KV Cache and Long Context (Track 1)
Monday, July 13 9:00 am–9:15 am
Opening Remarks and Awards
Monday, July 13 9:15 am–10:15 am
Keynote Address

Company intelligence

What does it mean when 70 companies show up, and 1 bet on three roles at once?

70 companies. one are betting more than once.

The numbers below are derived from the speakers, sponsors, and exhibitors on this page, cross-referenced into one ledger. They are the only thing here that 10times.com cannot tell you.

Who's in the room

1All threespeak · spons · exh

9Sponsoringsponsor only

60Speakingspeaker only

In all three roles

Microsoft

Speakers by seniority

Other roles146100%

Of 146 speakers · classified by free-text title

Most-represented companies

#	Company	Roles	Speakers
1	Shanghai Jiao Tong University		14
2	Georgia Institute of Technology		10
3	DistriNet - KU Leuven		7
4	Purdue University		6
5	Ruhr University Bochum		6
6	EPFL		5
7	Graz University of Technology		5
8	University of Birmingham		5
9	Worcester Polytechnic Institute		5
10	Alibaba Group		4
11	Microsoft	Sponsor · Bronze	1
12	NVIDIA		4
13	UCLA		4
14	Amazon	Sponsor · Platinum	-

Sponsors by tier

platinum1

10% of 10

gold3

30% of 10

silver3

30% of 10

bronze2

20% of 10

open access1

10% of 10

At peer events but not here

These companies sponsored two or more peer events recently, but aren't on this program. For an organizer, that's a list of warm prospects.

Google×93 peers
IBM×85 peers
Adobe×70 peers
Alibaba×69 peers
Facebook×69 peers
Snapchat×68 peers
Taboola×68 peers
HubSpot×67 peers
ManyChat×67 peers
YouTube×67 peers
Outbrain×65 peers
LinkedIn×64 peers

Related events · by shared sponsors

Conference	Date	City	Shared sponsors
MLSys 2027	June 21, 2027	Bellevue	5
Web Summit 2026	November 9, 2026	Lisbon	4
NRF Big Show 2027	January 10, 2027	New York	4
Affiliate Summit West 2027	January 26, 2027	Las Vegas	3
Affiliate Summit East 2027	July 26, 2027	New York	3
Privacy Conference 2027	September 28, 2027	Berlin	3

OSDI '26

Chimera: Cost-Efficient Multi-LLM Serving via GPU Memory Ballooning

Contextra: Hierarchical Context Caching for Long Context Language Model Serving

ECHO: Efficient KV Cache Offloading with Lossless Prefetching for Serving Native Sparse Attention LLMs

No Buffer, No Bottleneck: Efficient Zero-Copy KV Cache Offloading for Long-Context LLMs

Simple is Better: Multiplication May Be All You Need for LLM Request Scheduling

Opening Remarks and Awards

Keynote Address

70 companies. one are betting more than once.