Arvind Krishnamurthy
Professor
Builds data center and Internet-scale systems, recently programmable networks and systems for machine learning.
Arvind Krishnamurthy's lab at UW works in AI Systems & Hardware. 34 papers since 2023, 26 at top venues, mostly MLSys and NSDI. 72 alumni, most often to Google and Microsoft. Applications close .
At a glance
- papers, 2023–2026
- 9/ year
- at T1 venues
- 76%
- PhD students
- 0
- MS students
- 0
- alumni listed
- 72
Papers by yearfrom the lab homepage, 2023–2026
- 9 papers2026 · T1 9 · collecting
- 8 papers2025 · T1 6
- 9 papers2024 · T1 7
- 8 papers2023 · T1 4
T1 venuesOtherVenues: MLSys, NSDI, SIGCOMM, ASPLOS, OSDI
Research focuspaper titles that use each topic's words
- data center0 papers
- programmable0 papers
- builds data0 papers
- center0 papers
- paul0 papers
Paper list
From the lab homepage.
20269 papers
- MLSys (T1)Demystifying the Mixture of Experts Serving Tax (opens in a new tab)
- NSDI (T1)FAST: An Efficient Scheduler for All-to-All GPU Communication (opens in a new tab)
- NSDI (T1)ForestColl: Throughput-Optimal Collective Communications on Heterogeneous Network Fabrics (opens in a new tab)
- MLSys (T1)NEST: Network- and Memory-Aware Device Placement for Distributed Deep Learning (opens in a new tab)
- EuroSys (T1)Reducing the GPU Memory Bottleneck with Lossless Compression for ML (opens in a new tab)
- NSDI (T1)RLBoost: Harvesting preemptible resources for cost-efficient reinforcement learning on LLMs (opens in a new tab)
- ASPLOS (T1)SG-IOV: Socket-Granular I/O Virtualization for SmartNIC-Based Container Networks (opens in a new tab)
- ICLR (T1)Tactic: Adaptive Sparse Attention with Clustering and Distribution Fitting for Long-Context LLMs (opens in a new tab)
- MLSys (T1)TeleRAG: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval (opens in a new tab)
20258 papers
- NSDI (T1)Efficient Direct-Connect Topologies for Collective Communications (opens in a new tab)
- MLSys (T1)FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving (opens in a new tab)
- NSDI (T1)High-Level Programming for Application Networks (opens in a new tab)
- SOSP (T1)IC-cache: Efficient large language model serving via in-context caching (opens in a new tab)
- OSDI (T1)NanoFlow: Towards Optimal Large Language Model Serving Throughput (opens in a new tab)
- HotOSRethinking RPC Communication for Microservices-based Applications (opens in a new tab)
- CoNextSwitchNIC: A Hybrid Architecture for Network Functions with Fast and Consistent Shared State (opens in a new tab)
- NSDI (T1)White-boxing RDMA with Packet-Granular Software Control (opens in a new tab)
20249 papers
- SIGCOMM (T1)An Architecture For Edge Networking Services (opens in a new tab)
- MLSys (T1)Atom: Low-bit Quantization for Efficient and Accurate LLM Serving (opens in a new tab)
- ASPLOS (T1)CC-NIC: a Cache-Coherent Interface to the NIC (opens in a new tab)
- HPDCEfficient all-to-all Collective Communication Schedules for Direct-connect Topologies (opens in a new tab)
- SIGCOMM (T1)Principles for Internet Congestion Management (opens in a new tab)
- MLSys (T1)Punica: Multi-tenant LoRA serving (opens in a new tab)
- SIGCOMM (T1)Relational Network Verification (opens in a new tab)
- FPGASuperNIC: An FPGA-Based, Cloud-Oriented SmartNIC (opens in a new tab)
- SIGCOMM (T1)Understanding the Host Network (opens in a new tab)
20238 papers
- SOSP (T1)A Cloud-Scale Characterization of Remote Procedure Calls (opens in a new tab)
- SOCCAnticipatory Resource Allocation for ML Training (opens in a new tab)
- HotNetsApplication Defined Networks (opens in a new tab)
- SOCCDissecting Overheads of Service Mesh Sidecars (opens in a new tab)
- OSDI (T1)eZNS: An Elastic Zoned Namespace for Commodity ZNS SSDs (opens in a new tab)
- SIGCOMM (T1)Host Congestion Control (opens in a new tab)
- HotNetsHow I Learned to Stop Worrying About CCA Contention (opens in a new tab)
- SIGCOMM (T1)Unleashing SmartNIC Packet Processing Performance in P4 (opens in a new tab)
Where alumni went72 listed, 57% to industry
Industry 28Academia 21
Most often to
- Microsoft
- Meta
Also to
- UC San Diego
- USC
- UC Berkeley
Apply to UWFall 2027
UW Allen School Computer Science PhD
- Round
- Single round
- Opens
- GRE
- not required
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- USUMD
Cunxi Yu
- Accelerators, chips & architecture
- USUCSD
Hadi Esmaeilzadeh
- Accelerators, chips & architecture
- USStanford
Kunle Olukotun
- Accelerators, chips & architecture
- USUC Berkeley
Zaijun Chen
- Accelerators, chips & architecture
- GBEdinburgh
Luo Mai
- Accelerators, chips & architecture
- CAToronto
Maryam Mehri Dehnavi
- Compilers & programming languages
Your notes
Save this lab to keep notes.
Data
Some numbers could not be verified:
- This site is shared with other labs, so counts may overlap.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
This lab shares a website with other labs. Member and alumni counts may include them.
Tiers are venue classifications, not lab rankings.
UWcloses