Abstract

Network forwarding state undergoes frequent changes, in batches of forwarding rule modifications at multiple switches. Installing or modifying a large number of rules is time con-suming given the performance limits of current programmable switches, which are also due to economical factors in addi-tion to technological ones. In this paper, we observe that a large network-state up-date typically consists of a set of sub-updates that are in-dependent of one another w.r.t. the traffic they affect, and hence sub-updates can be installed in parallel, in any order. Leveraging this observation, we treat update installation as a scheduling problem and design ESPRES, a runtime mech-anism that rate-limits and reorders updates to fully utilize processing capacities of switches without overloading them. Our early results show that compared to using no scheduler, our schemes yield 2.17-3.88 times quicker sub-update com-pletion time for 20th percentile of sub-updates and 1.27-1.57 times quicker for 50th percentile.