On the efficacy of fine-grained traffic splitting protocols in data center networks

Dixit, Advait; Prakash, Pawan; Kompella, Ramana Rao; Hu, Charlie

doi:10.1145/2254756.2254818

Cited by 5 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ECMP is a typical switch-only routing protocol for load balance. Many existing switch-only protocols allow a flow take multiple paths at the same time (called flow splitting) to achieve high throughput [15,9,5]. Flow splitting may cause a high level of TCP packet reordering, resulting in throughput drop [16].…”

Section: Accepted Manuscriptmentioning

confidence: 99%

DiFS: Distributed Flow Scheduling for adaptive switching in FatTree data center networks

Cui

Qian

2016

Computer Networks

View full text Add to dashboard Cite

Data center networks leverage multiple parallel paths connecting end host pairs to offer high bisection bandwidth for cluster computing applications.However, the state-of-the-art routing protocols such as Equal Cost Multipath (ECMP) is load-oblivious due to the static flow-to-link assignment. They may cause bandwidth loss due to flow collisions. Recently proposed centralized scheduling algorithm or host based adaptive routing that requires network-wide state information may suffer from scalability problems. In this paper, we present Distributed Flow Scheduling (DiFS), a new adaptive switching method, for FatTree data center networks, which is a localized and switch-only solution. DiFS allows switches to cooperate to avoid overutilized links and find available paths without centralized control. DiFS is scalable and can react quickly to dynamic traffic because it is independently executed on switches and requires no synchronization. DiFS provides global bounds of flow balance based on local optimization. Extensive simulations show that the aggregate throughput of DiFS using various traffic patterns is much better than that of ECMP, and is similar to or higher than those of two representative protocols that use network-wide optimization.

show abstract

Section: Accepted Manuscriptmentioning

confidence: 99%

DiFS: Distributed Flow Scheduling for adaptive switching in FatTree data center networks

Cui

Qian

2016

Computer Networks

View full text Add to dashboard Cite

show abstract

“…Most of those sophisticated traffic engineering techniques manipulate route via changing the link weights and coupling with link state protocols like OSPF and ISIS. While they naturally fit ISP networks with nearly random topologies and high-end routers, they may not be good options for datacenters with relatively regular topologies and commodity switches, where people usually deploy simple spanning tree forwarding and ECMP [12]. Furthermore, many recently-proposed datacenters such as BCube [13], DCell [14], PortLand [15], etc.…”

Section: A Traffic Engineering In Datacentermentioning

confidence: 99%

VirtualKnotter: Online virtual machine shuffling for congestion resolving in virtualized datacenter

et al. 2014

View full text Add to dashboard Cite

Abstract-Our measurements on production datacenter traffic together with recently-reported results [1] suggest that datacenter networks suffer from long-lived congestion caused by core network oversubscription and unbalanced workload placement. In contrast to traditional traffic engineering approaches that optimize flow routing, in this paper, we explore the opportunity to address the continuous congestion via optimizing VM placement in virtualized datacenters. To this end, we present VirtualKnotter, an efficient online VM placement algorithm to reduce congestion with controllable VM migration traffic as well as low time complexity. Our evaluation with both real and synthetic traffic patterns shows that VirtualKnotter performs close to the baseline algorithm in terms of link unitization, with only 5%-10% migration traffic of the baseline algorithm. Furthermore, VirtualKnotter decreases link congestion time by 53% for the production datacenter traffic.

show abstract

“…While packet level ECMP has been proposed [30], it is not widely used. Second, DIBS spreads packets based on network load, not path length.…”

Section: Related Workmentioning

confidence: 99%

Dibs

Zarifis

Miao

Calder

et al. 2014

Proceedings of the Ninth European Conference on Computer Systems

View full text Add to dashboard Cite

Data centers must support a range of workloads with differing demands. Although existing approaches handle routine traffic smoothly, intense hotspots-even if ephemeralcause excessive packet loss and severely degrade performance. This loss occurs even though congestion is typically highly localized, with spare buffer capacity at nearby switches. In this paper, we argue that switches should share buffer capacity to effectively handle this spot congestion without the monetary hit of deploying large buffers at individual switches. Specifically, we present detour-induced buffer sharing (DIBS), a mechanism that achieves a near lossless network without requiring additional buffers at individual switches. Using DIBS, a congested switch detours packets randomly to neighboring switches to avoid dropping the packets. We implement DIBS in hardware, on software routers in a testbed, and in simulation, and we demonstrate that it reduces the 99th percentile of delay-sensitive query completion time by up to 85%, with very little impact on other traffic.

show abstract

On the efficacy of fine-grained traffic splitting protocols in data center networks

Cited by 5 publications

References 3 publications

DiFS: Distributed Flow Scheduling for adaptive switching in FatTree data center networks

DiFS: Distributed Flow Scheduling for adaptive switching in FatTree data center networks

VirtualKnotter: Online virtual machine shuffling for congestion resolving in virtualized datacenter

Dibs

Contact Info

Product

Resources

About