Dmitrii Ustiugov

NEWS: I am on the academic job market this year (2021-2022).  Potential employers can see my CV and contact me by email.

I am a postdoctoral researcher at ETH Zurich, received a PhD degree in Computer Science from the University of Edinburgh where was co-advised by Professors Boris Grot and Edouard Bugnion (EPFL). The leader of the vHive open-source framework & ecosystem for serverless experimentation.

My research interests span Computer Architecture and Computer Systems with a current focus on serverless and cloud computing as well as systems for machine learning.

Join the vHive Open-Source Community

vHive is a full-stack open-source framework & ecosystem for serverless experimentation and innovation. vHive is representative of real serverless clouds, as recognized by 2 leading serverless providers and an ASPLOS'21 Distinguished Artifact Award, integrating production-grade components, including AWS FirecrackerContainerd and Kubernetes.

Today, vHive is in use in 15+ top universities around the globe, for both research and teaching, supported and/or sponsored by 5 international IT companies.

Research Angles

There are three principal means of acquiring knowledge available to us: observation of nature, reflection, and experimentation. Observation collects facts; reflection combines them; experimentation verifies the result of that combination.
Denis Diderot, On the Interpretation of Nature


Serverless & Public Cloud

Fuse extreme multi-tenancy with high isolation guarantees for high performance and cost efficiency without security compromises.


Hybrid-Memory Server Architectures

Cut costs and boost speed by alloying storage-class memory with die-stack memory technologies in hybrid DIMM form-factor.


Disaggregated Servers for Memory Pooling

Leverage operating systems and architectural support for building low latency and high throughput disaggregated systems.


There is a time for many words, and there is also a time for sleep.
Homer, The Odyssey


Serverless & Public Cloud
Lukewarm Serverless Functions: Characterization and Optimization
D. Schall, A. Margaritov, D. Ustiugov, A. Sandberg, B. Grot
To appear in ISCA 2022

Analyzing Tail Latency in Serverless Clouds with STeLLAR
D.Ustiugov, T.Amariucai, B.Grot
IISWC 2021

Benchmarking, Analysis, and Optimization of Serverless Function Snapshots
D.Ustiugov, P.Petrov, M.Kogias, E.Bugnion, B.Grot
Distinguished Artifact Award

PTEMagnet: Fine-Grained Physical Memory Reservation for Faster Page Walks in Public Clouds
A.Margaritov, D.Ustiugov, A.Shahab, B.Grot

Bankrupt Covert Channel: Turning Network Predictability into Vulnerability
D.Ustiugov, P.Petrov, M.R.S.Katebzadeh, B.Grot
WOOT co-located with USENIX Security 2020

Prefetched Address Translation
A.Margaritov, D.Ustiugov, E.Bugnion, B.Grot
MICRO 2019

Virtual Address Translation via Learned Page Table Indexes
A.Margaritov, D.Ustiugov, E.Bugnion, B.Grot
NeurIPS, ML for Systems workshop 2018


Hybrid-Memory Server Architectures

Design Guidelines for High-Performance SCM Hierarchies
D.Ustiugov, A.Daglis, J.Picorel, M.Sutherland, E.Bugnion, B.Falsafi, D.Pnevmatikatos

Algorithm/Architecture Co-Design for Near-Memory Processing
M.Drumond, A.Daglis, N.Mirzadeh, D.Ustiugov, J.Picorel, B.Falsafi, B.Grot, D.Pnevmatikatos
ACM SIGOPS Operating Systems Review, 2018

The Mondrian Data Engine
M.Drumond, A.Daglis, N.Mirzadeh, D.Ustiugov, J.Picorel, B.Falsafi, B.Grot, D.Pnevmatikatos
ISCA 2017 


Disaggregated Servers for Memory Pooling

Mitigating load imbalance in distributed data serving with rack-scale memory pooling
S.Novakovic, A.Daglis, D.Ustiugov, E.Bugnion, B.Falsafi, B.Grot
TOCS 2019

SABRes: Atomic Object Reads for In-Memory Rack-Scale Computing
A.Daglis, D.Ustiugov, S.Novakovic, E.Bugnion, B.Falsafi, B.Grot
MICRO 2016

Teaching & Supervision

It is the supreme art of the teacher to awaken joy in creative expression and knowledge.
Albert Einstein

  • Extreme Computing. Invited lecturer, Fall 2021. University of Edinburgh. [Slides]
  • Operating Systems. Invited lecturer, Spring 2021. University of Edinburgh. [Slides]
  • Introduction to Computer Systems, Teaching assistant, Fall 2019, 2020, University of Edinburgh.
  • Extreme Computing, Teaching assistant, Fall 2019, University of Edinburgh.
  • Computer Architecture, Teaching assistant, Fall 2015, 2017, EPFL.
  • Statistics, Teaching assistant, Spring 2016, EPFL.
  • Introduction to Multiprocessor Architecture, Teaching assistant, Spring 2015, EPFL.
  • Shyam Jesalpura, Research Intern, BITS Pilani, 2021.
    Project: High-speed communication fabric for serverless clouds.
  • Mert Bora Alper, Undergraduate Intern, University of Edinburgh, 2021.
    Project: Benchmarking methodology for serverless clouds.
  • Michal Baczun, Undergraduate Intern, University of Edinburgh, 2021.
    Project: Representative suite of serverless workloads. 
  • Yuchen Niu, Undergraduate Student, University of Edinburgh, 2020-2021.
    Project: Implications of multi-tenancy on serverless hosts.
  • Theodor Amariucai, Undergraduate Student, University of Edinburgh, 2020-2021.
    Project: Tail latency analysis framework for serverless clouds.
  • Plamen Petrov, Undergraduate Student & Intern, University of Edinburgh, 2020.
    Project 1: End-to-end serverless benchmarking framework.
    Project 2: RDMA networks security and covert communication.
  • Ivy Wang, Undergraduate Student, University of Edinburgh, 2019.
    Project: Software support for contiguous page table allocation.
  • Sean Mullan, Undergraduate Student, University of Edinburgh, 2019.
    Project: Design exploration for TLB prefetching.
  • Lei Yan, Master Student, EPFL and RWTH Aachen University, 2018.
    Project: Design space exploration of cooperative scheduling for latency-critical cloud services.
  • Siddharth Gupta, PhD Student, EPFL, 2017.
    Project: Analysis of persistent memory systems on modern CPUs.
  • Nikhil Gupta, Undergraduate Intern, EPFL and IIT Delhi, 2016.
    Project: Robust infrastructure for QFlex (Flexus) simulation framework.
  • Virgile Neu, Undergraduate Student, EPFL, 2016.
    Project: Analyzing CPU front-end efficiency using perf counters. 

Collaboration & Funding

I thank my past and present collaborators round the globe for their professional and financial support.

  1. Oracle Labs generously provided one-year (2018) gift funding for the joint "Disaggregated processors" project at EPFL based on the research proposal I wrote with the help of Prof. Alex Daglis and Prof. Babak Falsafi.
  2. Google generously provided one-year Faculty Award (2019) for "Neural Address Translation" project at University of Edinburgh based on the research proposal written together with Prof. Boris Grot and my colleague Artemiy Margaritov.
  3. Arm generously provided 3-year fellowship at the University of Edinburgh, supporting our work on "Virtual Memory Software-Hardware Co-Design for Cloud Computing".

Professional Activities

  1. MICRO'22 TPC member.
  2. The head organizer of the Serverless & vHive Tutorial, co-located with ASPLOS'22.
  3. External reviewer at ATC'19, ISCA'19, CAL'19, 20; MDPI JLPEA'22, MDPI Electronics'22.
  4. EuroSys'21 ShadowPC member.
  5. SOSP'21 session co-chair.
  6. ASPLOS'20 Artifact Evaluation Committee member.

Awards and Recognition

© Copyright 2021 Dmitrii Ustiugov - All Rights Reserved
Email: dmitrii(dot)ustiugov(at)

Built with Mobirise - See here