What is Hive?
(click
here
for our SOSP'95 paper)
Reliability and scalability are major concerns when designing operating systems for large-scale shared-memory multiprocessors. Hive is an operating system with a novel kernel architecture that addresses these issues. Hive is structured as an internal distributed system of independent kernels called cells. This improves reliability because a hardware or software fault damages only one cell rather than the whole system, and improves scalability because few kernel resources are shared by processes running on different cells. The Hive prototype is a complete implementation of UNIX SVR4 and is targeted to run on the Stanford FLASH multiprocessor.Associated with the design of Hive is the development of new simulation technologies. In order to design and evaluate an operating system on hardware years from completion, simulation technology is a must. The underlying goal of our simulation work is to provide multiple levels of simulation detail. The different levels of detail can be switched between dynamically, allowing a researcher to "position" their workloads with a fast simulator, and only switch into slower, more detailed levels at the more interesting points of an execution.
For a detailed discussion of those topics, please browse the hypertext version of our SOSP'95 paper on Hive.
The Hive OS Team
Publications
SimOS Home Page
(simulation environment)
Publications
