BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:PhD Thesis Defense: Ethan Johnson\, Computer Science
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20261015T020632Z
UID:tag:localist.com\,2008:EventInstance_53506703533129
DTSTART:20260811T180000Z
DTEND:20260811T190000Z
DESCRIPTION:Ethan Johnson\, "Securing and Accelerating Hypervisors with Vir
 tual Instruction Set Computing"\n\nAdvisor: Prof. John Criswell (Computer 
 Science)  \n\nCommittee: Prof. Chen Ding (Computer Science)\, Prof. Michae
 l L. Scott (Computer Science)\, Prof. Emmett Witcher (UT Austin)\n\nChair:
  William R. Donaldson (Laboratory for Laser Energetics)\n\nModern cloud co
 mputing promises a compelling bargain: by virtualizing and containerizing 
 our digital application workloads\, we can safely and efficiently enable m
 utually distrustworthy tenants to share the same physical hardware by exte
 nding the same time-sharing techniques that have long been the foundation 
 of multitasking operating systems. The widespread deployment and adoption 
 of virtualization has delivered massive gains in economies of scale\, modu
 larity\, and resiliency\, enabling even the smallest tenants to fluidly ac
 cess tailored yet staggeringly powerful computing resources while benefiti
 ng from levels of redundancy and operational assurance that were once the 
 exclusive province of large enterprises with full-time IT divisions. Today
 \, even hobbyists and undergraduate students routinely deploy network appl
 ications to public clouds like Amazon Web Services (AWS) and Microsoft Azu
 re\, enjoying their own private slice of enterprise-level datacenter infra
 structure for a few dollars a month.\n\nThis entire concept is premised on
  the idea that virtualized\, shared systems can provide strong security bo
 undaries between tenants. In modern clouds\, the lion's share of that secu
 rity responsibility is placed upon the hypervisor\, the software layer res
 ponsible for providing each tenant with a virtualized view of a complete c
 omputer system\, called a "virtual machine" (VM). The hypervisor mediates 
 between those VMs and the physical hardware on which they run\, ensuring t
 he critical security dimensions of confidentiality\, integrity\, and avail
 ability are upheld between VMs even as they "live in the same brain"\, com
 putationally speaking.\n\nBut hypervisors\, like operating system kernels 
 before them\, are ultimately just software---complex\, intricate works of 
 logic\, vulnerable to all manner of mistakes and abstraction violations th
 at can undermine the security guarantees they are supposed to uphold. Crit
 ical hypervisor vulnerabilities are discovered and exploited every year\, 
 compromising the valuable data and responsibilities with which users entru
 st these systems and shaking confidence in the feasibility of deploying se
 nsitive workloads in public clouds. Many of these vulnerabilities occur at
  the lowest levels of software abstraction\, calling for solutions that un
 derstand and address the semantic gaps that arise from their nuanced inter
 actions with the computer's hardware and other software.\n\nPrior work has
  shown that virtual instruction set computing (VISC) techniques are a high
 ly effective way to implement strong security hardening in low-level\, pri
 vileged-mode operating system software with competitive performance tradeo
 ffs. This dissertation posits\, and demonstrates\, the thesis that such te
 chniques can be applied effectively to virtual machine hypervisors\, and i
 n fact tend to yield much more favorable security/performance tradeoffs in
  this context due to the high trust and small runtime fraction required of
  hypervisors compared to operating system kernels. Furthermore\, these tec
 hniques can be used not only to add hardening to existing threat models\, 
 but also to supplant existing hardware-based isolation relied upon by hype
 rvisors—and in some contexts\, this has the potential to dramatically im
 prove performance compared to the conventional non-VISC approaches in comm
 on use.
GEO:43.126069;-77.629191
LOCATION:Wegmans Hall\, 2506
SUMMARY:PhD Thesis Defense: Ethan Johnson\, Computer Science
URL;VALUE=URI:https://events.rochester.edu/event/phd-thesis-defense-ethan-j
 ohnson-computer-science
END:VEVENT
END:VCALENDAR