Karla Saur

I am currently a distributed systems engineer at Nvidia DGX Cloud, working with multi-cloud Kubernetes environments.


Summary:

My current interests are optimizing and automating cloud infrastructure for AI and machine learning workloads. I help create innovative solutions that enhance scalability, efficiency, and availability, particularly in containerized and serverless systems.

Professional Experience:
  • Prior to joining Nvidia, I was a Principal Research SDE in Gray Systems Lab (GSL) at Microsoft. There, I researched how to optimize containerized infrastructure [1] [2] and co-created Hummingbird, a library for compiling trained traditional ML models into tensor computations.
  • I have also worked on the Azure Kubernetes Service (AKS) infrastructure team and at Intel Labs as a research scientist focusing on distributed systems.
  • During my PhD at the University of Maryland College Park, I studied dynamic software updates for systems requiring high availability. My dream was (and is!) to eliminate all downtime in running systems.
  • I started my career in the information security field in the Maryland/DC area.

Publications

  • Vertically Autoscaling Monolithic Applications with CaaSPER: Scalable Container-as-a-Service Performance Enhanced Resizing Algorithm for the Cloud
    Anna Pavlenko, Joyce Cahoon, Yiwen Zhu, Brian Kroth, Michael Nelson, Andrew Carter, David Liao, Travis Wright, Jesús Camacho Rodríguez, Karla Saur. SIGMOD, 2024. [pdf]
  • VASIM: Vertical Autoscaling Simulator Toolkit
    Anna Pavlenko, Karla Saur, Yiwen Zhu, Brian Kroth, Joyce Cahoon, Jesús Camacho Rodríguez. ICDE, 2024. [pdf]
  • Containerized Execution of UDFs: An Experimental Evaluation
    Karla Saur, Tara Mirmira, Konstantinos Karanasos, Jesús Camacho Rodríguez. VLDB, 2022. [pdf]
  • Query Processing on Tensor Computation Runtimes
    Dong He, Supun C Nakandala, Dalitso Banda, Rathijit Sen, Karla Saur, Kwanghyun Park, Carlo Curino, Jesús Camacho Rodríguez, Konstantinos Karanasos, Matteo Interlandi. VLDB, 2022. [pdf]
  • End-to-end Optimization of Machine Learning Prediction Queries
    Kwanghyun Park, Karla Saur, Dalitso Banda, Rathijit Sen, Matteo Interlandi, Konstantinos Karanasos. SIGMOD, 2022. [pdf]
  • Tensors: An abstraction for general data processing
    Dimitrios Koutsoukos, Supun Nakandala, Konstantinos Karanasos, Karla Saur, Gustavo Alonso, Matteo Interlandi. VLDB, 2021. [pdf]
  • A Tensor Compiler for Unified Machine Learning Prediction Serving
    Supun Nakandala, Karla Saur, Gyeong-In Yu, Konstantinos Karanasos, Carlo Curino, Markus Weimer, Matteo Interlandi. OSDI, 2020. [pdf]
  • Cloudy with high chance of DBMS: A 10-year prediction for Enterprise-Grade ML
    Ashvin Agrawal, Rony Chatterjee, Carlo Curino, Avrilia Floratou, Neha Gowdal, Matteo Interlandi, Alekh Jindal, Kostantinos Karanasos, Subru Krishnan, Brian Kroth, Jyoti Leeka, Kwanghyun Park, Hiren Patel, Olga Poppe, Fotis Psallidas, Raghu Ramakrishnan, Abhishek Roy, Karla Saur, Rathijit Sen, Markus Weimer, Travis Wright, Yiwen Zhu. CIDR, 2020. [pdf]
  • Evolving NoSQL Databases Without Downtime
    Karla Saur, Tudor Dumitraș, and Michael Hicks. International Conference on Software Maintenance and Evolution (ICSME), October 2016. [pdf]
  • Safe and Flexible Controller Upgrades for SDNs
    Karla Saur, Joseph Collard, Nate Foster, Arjun Guha, Laurent Vanbever, and Michael Hicks. Symposium on SDN Research (SOSR), March 2016. [pdf]
  • C-strider: Type-Aware Heap Traversal for C
    Karla Saur, Michael Hicks, and Jeffrey S. Foster. Software: Practice and Experience, 2015. DOI: 10.1002/spe.2332 [pdf]
  • Kitsune: Efficient, General-purpose Dynamic Software Updating for C
    Christopher M. Hayden, Karla Saur, Edward K. Smith, Michael Hicks, and Jeffrey S. Foster. ACM Transactions on Programming Languages and Systems (TOPLAS), Vol. 36, No. 4, Article 13, Publication date: October 2014. [pdf]
  • A Study of Dynamic Software Update Quiescence for Multithreaded Programs
    Christopher M. Hayden, Karla Saur, Michael Hicks, and Jeffrey S. Foster, In Proceedings of the Workshop on Hot Topics in Software Upgrades (HotSWUp), June 2012. [pdf] [slides]
  • Locating x86 Paging Structures in Memory Images
    Karla Saur, Julian B. Grizzard, Journal of Digital Investigation, Vol. 7, pp. 28-37, 2010. [pdf] [slides]

Patents (Granted)

  • Query processing with machine learning.
    2025/3/4, US Patent 12,242,493 [pdf]
  • Technologies for providing guidance for autonomous vehicles in areas of low network connectivity.
    2023/1/10, US Patent 11,551,551 [pdf]
  • Methods and apparatus to collect and analyze rating information.
    2022/4/19, US Patent 11,308,510 [pdf]
  • Scaling mobile gateways in a 3rd generation partnership project (3GPP) network.
    2020/3/24, US Patent 10,602,349 [pdf]
  • Technology for secure partitioning and updating of a distributed digital ledger.
    2020/1/21, US Patent 10,540,652 [pdf]