BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards Grey Fault Tolerant Cloud Systems - Ryan Huang
DTSTART:20210401T140000Z
DTEND:20210401T150000Z
UID:TALK157828@talks.cam.ac.uk
CONTACT:Amjad
DESCRIPTION:Building robust\, large-scale distributed systems is notorious
 ly\nchallenging. Decades of research have made significant advances in\nta
 ckling this challenge with mature techniques such as state-machine\nreplic
 ation. These techniques usually assume a fail-stop model. Ample\nreal-worl
 d evidence\, however\, suggests that faults in modern cloud\ninfrastructur
 e are often "grey"\, in which a component is severely\nimpaired but still 
 appears to be working. These grey failures cannot\nbe effectively detected
  or handled by existing solutions.\n\nIn this talk\, I will discuss the gr
 ey failure problem. Using\nreal-world examples\, we argue that a key trait
  of the subtle grey\nfailure mode is a form of differential observability.
  Based on this\ninsight\, I will present Panorama\, a solution that harnes
 ses\nobservability in large systems to detect grey failures by using\ninst
 rumentation to convert any system component into an in-situ\nobserver. To 
 further enhance the inherent system observability\, I will\npropose an int
 rinsic software watchdog abstraction and a tool called\nOmegaGen that auto
 matically generates customized watchdogs for a given\nprogram by using a p
 rogram reduction technique. I will conclude by\noutlining some open challe
 nges in making cloud systems\ngrey-fault-tolerant.\n\nBio:\n\nRyan Huang i
 s an Assistant Professor in the Department of Computer\nScience at Johns H
 opkins University. He leads the Ordered Systems Lab\nat JHU\, which conduc
 ts research broadly in distributed systems\,\noperating systems\, cloud an
 d mobile computing. His work received the\nbest paper award at OSDI 2016\,
  ASPLOS 2019\, NSDI 2020\, and the best\npaper award nominee at MICRO 2018
 . He is a recipient of the NSF CAREER\nAward (2020). Dr. Huang received a 
 B.S. degree in Computer Science\n(Economics minor) from Peking University 
 (2010)\, a P.h.D degree from\nUC San Diego (2016).
LOCATION:meet.google.com/rvx-amuv-xrh
END:VEVENT
END:VCALENDAR