{
  "id": "1512.07771",
  "version": "v1",
  "published": "2015-12-24T09:52:19.000Z",
  "updated": "2015-12-24T09:52:19.000Z",
  "title": "Achievable Performance of Blind Policies in Heavy Traffic",
  "authors": [
    "Nikhil Bansal",
    "Bart Kamphorst",
    "Bert Zwart"
  ],
  "categories": [
    "math.PR",
    "cs.DS"
  ],
  "abstract": "For a GI/GI/1 queue, we show that the average sojourn time under the (blind) Randomized Multilevel Feedback algorithm is no worse than that under Shortest Remaining Processing Time algorithm times a logarithmic function of the system load. Moreover, it is verified that this bound is tight in heavy traffic, up to a constant multiplicative factor. We obtain this result by combining techniques from two disparate areas: competitive analysis and applied probability.",
  "revisions": [
    {
      "version": "v1",
      "updated": "2015-12-24T09:52:19.000Z"
    }
  ],
  "analyses": {
    "keywords": [
      "heavy traffic",
      "blind policies",
      "achievable performance",
      "remaining processing time algorithm times",
      "shortest remaining processing time algorithm"
    ],
    "note": {
      "typesetting": "TeX",
      "pages": 0,
      "language": "en",
      "license": "arXiv",
      "status": "editable"
    }
  }
}