A price-aware congestion control protocol for cloud services

Abstract

In current infrastructure-as-a service (IaaS) cloud services, customers are charged for the usage of computing/storage resources only, but not the network resource. The difficulty lies in the fact that it is nontrivial to allocate network resource to individual customers effectively, especially for short-lived flows, in terms of both performance and cost, due to highly dynamic environments by flows generated by all customers. To tackle this challenge, in this paper, we propose an end-to-end Price-Aware Congestion Control Protocol (PACCP) for cloud services. PACCP is a network utility maximization (NUM) based optimal congestion control protocol. It supports three different classes of services (CoSes), i.e., best effort service (BE), differentiated service (DS), and minimum rate guaranteed (MRG) service. In PACCP, the desired CoS or rate allocation for a given flow is enabled by properly setting a pair of control parameters, i.e., a minimum guaranteed rate and a utility weight, which in turn, determines the price paid by the user of the flow. Two pricing models, i.e., a coarse-grained VM-Based Pricing model (VBP) and a fine-grained Flow-Based Pricing model (FBP), are proposed. The optimality of PACCP is verified by both large scale simulation and small testbed implementation. The price-performance consistency of PACCP are evaluated using real datacenter workloads. The results demonstrate that PACCP provides minimum rate guarantee, high bandwidth utilization and fair rate allocation, commensurate with the pricing models.

Adaptable L4S Congestion Control for Cloud-Based Real-Time Streaming Over 5G

“…CC has been a crucial research topic since 1988 [13], and numerous CCs in Transmission Control Protocol (TCP) have been proposed: the window-based CC like TCP-Vega [14], TCP-Cubic [15], and Copa [16] adapt the window size derived from the congestion metrics; the rate-based CC like PCC [17], TCP-BBR [18] adjust the transmission rate for fast re-synchronization to reduce frame delay. Recently, CC in TCP specialized for cloud services has been proposed [19] [20].…”

“…We utilize Jain's fairness [23], measured instantaneously at each 100ms interval, to evaluate fairness. For all cases, Table 1 presents link utilization, QD percentile 95 % and 99 %, TDQ (20), TDQ(55), TDQ(145), and Jain's fairness index on average of two traffics on IETF pattern. GCC performs to achieve fair distribution, with Jain's fairness values ranging from 0.97 to 0.98, similar to those of cL4S and sL4S.…”