News

• October 2021

  Predicting Network Buffer Capacity for BBR Fairness paper got accepted NeurIPS MLSys Workshop

• September 2021

  Improving Storage Systems Using Machine Learning paper got accepted ACM TOS

• June 2021

  KML paper got accepted HotStorage 2021

• 
  June 2021

  I started doing internship at Facebook

• April 2020

  Re-Animator paper got accepted SYSTOR 2020

• February 2020

  KMLib paper got accepted On-device Intelligence Workshop at MLSys

• 
  June 2019

  I started doing internship in vSAN team at VMware

• 
  May 2018

  I started doing internship in File System team at Datrium (update: acquired by VMware)

About Me

I am a PhD. student at the Stony Brook University in the File System Lab (FSL) group. I am working with Prof. Erez Zadok. Before joining FSL, I have worked with Prof. Mike Ferdman at COMPAS Lab.

Research Interests

I am mainly interested in Operating Systems, File Systems, Storage Systems, Compilers, and Computer Architecture. One of my main goals is building fully adaptive self-driven systems software. I have been working on building scalable and verifiable system benchmarking.

Projects

You can look at all current and finished projects at Projects page. I am currently working on:

• KML: Bringing Machine Learning to Linux Kernel.
• Re-Animator: High quality tracing is important.

Publications

2022

• Predicting Network Buffer Capacity for BBR Fairness
  Ibrahim Umit Akgun, Santiago Vargas, Michael Arkhangelskiy, Andrew Burford, Michael McNeill, Aruna Balasubramanian, Anshul Gandhi, and Erez Zadok
  NeurIPS MLSys Workshop 2022, To Appear,
  Paper | Bibtex

• Improving Storage Systems Using Machine Learning
  Ibrahim Umit Akgun, Ali Selman Aydin, Andrew Burford, Michael McNeill,Michael Arkhangelskiy, Erez Zadok
  ACM Transactions on Storage (TOS) 2022, To Appear,
  Paper | Bibtex | KML Source Code

2021

• A Machine Learning Framework to Improve Storage System Performance
  Ibrahim Umit Akgun, Ali Selman Aydin, Aadil Shaikh, Lukas Velikov, Erez Zadok
  Proceedings of the 13^th ACM Workshop on Hot Topics in Storage and File Systems (HotStorage'21),
  Paper | Bibtex | Slides

2020

• Re-Animator: Versatile High-Fidelity Storage-System Tracing and Replaying
  Ibrahim Umit Akgun, Geoff Kuenning, Erez Zadok
  Proceedings of the 13^th ACM International System and Storage Conference (SYSTOR'20),
  Paper | Bibtex | Slides | Presentation | Re-Animator LTTng

• KMLib: Towards Machine Learning For Operating Systems
  Ibrahim Umit Akgun, Ali Selman Aydin, Erez Zadok
  On-device Intelligence Workshop - Proceedings of the 3^rd MLSys Conference, Austin, TX, USA,
  Paper Bibtex

• KMLib: Towards Machine Learning For Operating Systems Storage Components
  Ibrahim Umit Akgun, Ali Selman Aydin, Erez Zadok
  18th USENIX Conference on File and Storage Technologies, USENIX FAST'20 — Poster
  Poster

2019

• Re-Animator: Versatile High-Fidelity System-Call Tracing and Replaying
  Ibrahim Umit Akgun, Erez Zadok
  Technical Report FSL 19-02 Research Proficiency Exam, Stony Brook University — May 2019
  Report Bibtex Slides | Presentation