The UNIX system has been in wide use for over 20 years, and has helped to define many areas of computing.
Post
Replies:
15
Boosts:
16
i think it can work
An interactive job, such as a text editor searching for a string, may become compute bound briefly,
btw this is literally exactly what i hope my parsing phd thesis would be around. declaring structured dependencies between "primitive" matching operations over dependent string regions
The system also needs a scheduling policy to deal with problems that arise from not having enough main memory to hold the execution contexts of all processes that want to execute.
"want to execute" YOU ARE THE OS!!!! STAND UP FOR YOURSELF!!!!!
Numerous papers have been written about alternative scheduling policies, such as those used in batch-processing environments or real-time systems.
NUMEROUS!
Usually, these policies require changes to the system in addition to alteration of the scheduling policy [Khanna et al, 1992].
YOU GOT IT!
of course the citation is about real-time scheduling
oh hell yeah https://sci-hub.st/10.1002/cpe.904
In this study, we restrict our attention to the scheduling subsystem, focusing on techniques for improving protocol performance (i.e. low latency and packet-loss) and CPU utilization in distributed computing environments.
literally yes
It is our belief that direct access to network interfaces by non-privileged user-space processes will be supported securely by most commercial operating systems in the near future.
everyone always talks about network shit. network is easy!
Furthermore, user-level threads are portable and do not require kernel modifications or kernel coding.
BOOOOOO!!!! QUITTER
hell fucking yeah https://sci-hub.st/10.1109/IPDPS53621.2022.00043
DFMan: A Graph-based Optimization of Dataflow
Scheduling on High-Performance Computing
Systems
team llnl dot gov killing the game
In particular, we devise a graph-based optimization algorithm that can leverage an intuitive graph representation of dataflow- and system-related information, and automatically carry out co-scheduling of task and data placement.
it's that fucking easy
Efficient data management in HPC workflows requires the development of not only optimized scheduling strategies for data placement, but also proper assignment of computation resources to tasks.
ok they assume DAG structure and whip out the linear programming which sounds great for HPC workloads like weather simulations but is as usual trying to solve a harder problem and not seeking to also provide strong memory isolation and/or persistence guarantees
LLNL was easily the best job i've ever had and now i'm wondering if this is more of a candidate for phd study than parse scheduling
i feel that using the dependency graph to schedule the setup/shutdown of a task-specific memory and resource environment for isolated process execution is a really unique approach and really worth pursuing
