文摘
We consider the maximum flow time minimization problem in batch scheduling, which is a capacitated version of broadcast scheduling. In this setting, d="mmlsi1" class="mathmlsrc">data-mathURL="/science?_ob=MathURL&_method=retrieve&_eid=1-s2.0-S0167637716301134&_mathId=si1.gif&_user=111111111&_pii=S0167637716301134&_rdoc=1&_issn=01676377&md5=4872d5e0cb57292568c90e473e0597bd" title="Click to view the MathML source">ndden">de"> different pages of information are available at the server which receives requests from clients over time for specific pages. The server can transmit at most one page d="mmlsi2" class="mathmlsrc">data-mathURL="/science?_ob=MathURL&_method=retrieve&_eid=1-s2.0-S0167637716301134&_mathId=si2.gif&_user=111111111&_pii=S0167637716301134&_rdoc=1&_issn=01676377&md5=f9c819017148d39f0bde8217f20d2eb9" title="Click to view the MathML source">pdden">de"> at each time to satisfy a batch of requests for the same page d="mmlsi2" class="mathmlsrc">data-mathURL="/science?_ob=MathURL&_method=retrieve&_eid=1-s2.0-S0167637716301134&_mathId=si2.gif&_user=111111111&_pii=S0167637716301134&_rdoc=1&_issn=01676377&md5=f9c819017148d39f0bde8217f20d2eb9" title="Click to view the MathML source">pdden">de">, up to a certain capacity d="mmlsi4" class="mathmlsrc">data-mathURL="/science?_ob=MathURL&_method=retrieve&_eid=1-s2.0-S0167637716301134&_mathId=si4.gif&_user=111111111&_pii=S0167637716301134&_rdoc=1&_issn=01676377&md5=6650eb6dff2bb8d1c757993b706d49c9" title="Click to view the MathML source">Bpdden">de">. In this paper we give the first d="mmlsi5" class="mathmlsrc">data-mathURL="/science?_ob=MathURL&_method=retrieve&_eid=1-s2.0-S0167637716301134&_mathId=si5.gif&_user=111111111&_pii=S0167637716301134&_rdoc=1&_issn=01676377&md5=a6f71710056ef44986d856cda4ebf5bf" title="Click to view the MathML source">(1+ϵ)dden">de">-approximations for this problem with arbitrarily small resource augmentation, using either more capacity or more speed.