Congratulations @_Sartakov @LluisVilanova_ Munir Geden David Eyers @shina_en @ppietzuch https://t.co/Vf2Wfc9jP3

By LSDSImperial, 1 year 2 months ago
  • RT @ppietzuch: Existing distributed #reinforcement learning (RL) systems lack abstractions to distribute algorithms flexibly. Check out @LSDSImperial's work on #MSRL, an RL system that supports distribution polices for partitioning computation using dataflow graphs. https://t.co/ZKLWh9sbm3

    By LSDSImperial, 1 year 7 months ago
  • RT @SysML_ICL: 🚨 Announcing our exciting July Seminar line-up!
    Follow the links for details.

    📅 14/07: Davis Blalock will present his work "Multiplying Matrices Without Multiplying", ICML'21
    (1/3)
    https://t.co/AHUmZOEnBQ

    By LSDSImperial, 1 year 10 months ago
  • Please join us Thursday (9/6) at 2 pm BST for Girish Balakrishnan, IITM, talk "Succinct Data Structure for Path Graphs" https://t.co/C8y0UXkaoK

    By LSDSImperial, 1 year 11 months ago
  • Please join us Thursday (26/5) at 2 pm BST for Chi Wang, Microsoft Research, talk "Chi Wang, Microsoft Research" https://t.co/Gv3QPXVhS4

    By LSDSImperial, 2 years 1 day ago
  • Please join us Thursday (19/5) at 3:15pm BST for Peter Boncz, Vrije Universiteit Amsterdam, talk "FSST: fast random-access string compression" https://t.co/r9rOZZdxRT

    By LSDSImperial, 2 years 1 week ago
  • Please join us Thursday (19/5) at 2pm BST for Heming Cui, University of Hong Kong, talk "Building Four-dimensional Parallel Training Systems for Large AI Models" https://t.co/321J27Pk07

    By LSDSImperial, 2 years 1 week ago
  • Please join us Thursday (12/5) at 2pm BST for Luis Rodrigues, Universidade de Lisboa, talk "Kauri: BFT Consensus with Pipelined Tree-Based Dissemination and Aggregation" https://t.co/eBgnOWpxS8

    By LSDSImperial, 2 years 2 weeks ago
  • Please join us Thursday (28/4) at 2pm GMT for Jiwon Seo, Hanyang University, talk "Out-Of-Order BackProp: An Effective Scheduling Technique for Deep Learning" https://t.co/CY4RPNEpQ8

    By LSDSImperial, 2 years 4 weeks ago
  • RT @ppietzuch: Excited about OS kernels, hypervisors and low-level systems programming? Do you want to figure out how to make future cloud computing stacks more secure? At @LSDSImperial, we have fully-funded #PhD Studentships for international students. Apply here: https://t.co/zNZivRGn8X

    By LSDSImperial, 2 years 1 month ago

The goal of the LSDS group is to support the design and implementation of tomorrow's large-scale data-intensive systems. We investigate new abstractions and infrastructures for building scalable, robust and secure data-intensive applications.

Current Projects

  • View All
  • Big Data Processing
  • Serverless
  • Cloud Computing
  • Systems Security
  • Machine Learning
  • Networking
BOSS
BOSS
A Bring-Your-Own-Kernel DBMS
CloudCAP
CloudCAP
Compartments and Cloud-Native Applications
Crossbow
Crossbow
Scalable Multi-GPU Deep Learning
Faasm
Faasm
High performance serverless computing
FractOS
FractOS
Secure and Efficient Disaggregated Systems
KungFu
KungFu
Adaptive Distributed Machine Learning
LightSaber
LightSaber
Multi-core Window-Based Stream Processing
Medea
Medea
Heterogenous Application Scheduling
SGX-LKL
SGX-LKL
Linux Binaries in SGX Enclaves
Teechain
Teechain
A Secure Blockchain Payment Network

Recent Publications

MSRL: Distributed Reinforcement Learning with Dataflow Fragments
H. Zhu; B. Zhao; G. Chen; W. Chen; Y. Chen; S. Liang; Y. Yang; P. Pietzuch; and L. Chen

In USENIX ATC'23

Translation Pass-Through for Near-Native Paging Performance in VMs
S. Bergman; M. Silberstein; T. Shinagawa; P. Pietzuch; and L. Vilanova

In USENIX ATC'23

ORC: Increasing Cloud Memory Density via Object Reuse with Capabilities
V. Sartakov; L. Vilanova; M. Geden; D. Eyers; T. Shinagawa; and P. Pietzuch

In USENIX OSDI'23

Scabbard: Single Node Fault-Tolerant Stream Processing
G. Theodorakis; F. Kounelis; P. Pietzuch; and H. Pirk

In VLDB'22

Upcoming Events

Different Scales of Resource-Aware Deep Learning & How to Tackle Them
Pınar Tözün
IT University of Copenhagen
Thursday, 6 June 2024 - 2:00pm, Imperial College London
Today, deep learning runs at various scales of hardware resources from the cloud and high-performance computing (HPC) centers to edge and Internet-of-Things (IoT) devices. To achieve resource-aware...
Mysticeti: The New Core of the Sui Blockchain
Alberto Sonnino
Mysten Labs
Thursday, 30 May 2024 - 2:00pm, Imperial College London
This talk introduces Mysticeti a byzantine consensus protocol with low-latency and high resource efficiency. It leverages a DAG based on Threshold Clocks and incorporates innovations in pipelining...

The LSDS Group

Marios Kogias
Marios Kogias
Assistant Professor
Peter Pietzuch
Peter Pietzuch
Professor
Holger Pirk
Holger Pirk
Associate Professor
Lluís Vilanova
Lluís Vilanova
Associate Professor
Munir Geden
Munir Geden
Post-doctoral Researcher
Andrea Piermarteri
Andrea Piermarteri
Post-doctoral Researcher
Vasily A. Sartakov
Vasily A. Sartakov
Post-doctoral Researcher
Huanzhou Zhu
Huanzhou Zhu
Post-doctoral Researcher
Guo Li
Guo Li
Research Assistant
Michael Paper
Michael Paper
Research Assistant
Jack Pearce
Jack Pearce
Research Assistant
Konstantinos Fertakis
Konstantinos Fertakis
PhD Student
Ahmad Khazale
Ahmad Khazale
PhD Student
Fotis Kounelis
Fotis Kounelis
PhD Student
Eleftheria Mappoura
Eleftheria Mappoura
PhD Student
Domagoj Margan
Domagoj Margan
PhD Student
Carlos Segarra
Carlos Segarra
PhD Student
Jinnan Guo
Jinnan Guo
PhD Student
Scofield Liu
Scofield Liu
PhD Student
Alessandro Fogli
Alessandro Fogli
PhD Student
Pedro Silvestre
Pedro Silvestre
PhD Student
Marcel Wagenländer
Marcel Wagenländer
PhD Student
David Loughlin
David Loughlin
PhD Student
Hubert Mohr-Daurat
Hubert Mohr-Daurat
PhD Student
Takahiro Shinagawa
Takahiro Shinagawa
Visitor
Yanda Tao
Yanda Tao
Research Assistant
MINGXUAN YANG
MINGXUAN YANG
Visitor

Open Positions

Post-Doctoral Researcher
In Distributed Systems, Systems, Databases
 

Multiple post-doctoral positions available in the areas of distributed systems, databases, and networking.

PhD position
In Computer Systems for BigData and Data Science
 

Two PhD positions are available in the broad area of Computer Systems with focus on designing and building systems that support BigData and Data Science workloads on modern and future hardware platforms.

PhD position
In Big Data Management: Auto-tuning for modern hardware
 

PhD position in Big Data Management with focus on Auto-tuning for modern hardware under the supervision of Dr. Holger Pirk.

PhD position
In in Big Data Management: Data Compression
 

PhD position in Big Data Management with focus on Data Compression under the supervision of Dr. Holger Pirk.

Interested in joining LSDS? How to apply