Please enable JavaScript.

Coggle requires JavaScript to display documents.

PhD (Scheduling (Core (Altruistic Scheduling in Multi-Resource Clusters (•…

- - - - • We propose an altruistic, long-term approach, CARBYNE, where jobs yield fractions of their allocated resources without impacting their own completion times
    - - 3Sigma schedules jobs based on full distributions of relevant runtime histories
    - - novel way in which short-lived critical OS tasks are migrated to dedicated micro-sliced cores, minimizing the delays caused by time sharing.
    - - leverages information supplied by the calendaring reservation system about jobs' deadlines and estimated runtimes to plan ahead in deciding whether to wait for a busy preferred resource type or fall back to less preferred placement options.
    - - high percentiles of the total processor usage (summed over at least 10 tasks) can be reasonably estimated by the Gaussian distribution. We use this result for a probabilistic fit test, for standard bin-packing algorithms
    - - multidimensional case is inherently different from the single dimension case. Two of our new heuristics that dynamically change the placement logic according to the amount of available (unused) resources are shown to perform very well in many practical scenarios
  - - - Cluster scheduler, GRAPHENE, aimed at jobs that have a complex dependency structure and heterogeneous resource demands. compute a DAG schedule, offlin
    - - introduces powerful placement constraints with formal semantics to capture interactions among containers within and across applications.
  - - - Firmament,is a centralized scheduler that scales to over ten thousand machines at sub-second placement latency even though it continuously reschedules all tasks via a min-cost max-flow (MCMF) optimization.
    - - Claims to be more scalable