Hedera: Dynamic Flow Scheduling for Data Center Networks

By Mohammad Al-Fares, Sivasankar Radhakrishnan, Barath Raghavan, Nelson Huang, Amin Vahdat. Symposium on Networked Systems Design and Implementation, 2010

1,725 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/19ff9dac013d1ebca1ea1c9845325c9ddafdf93a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

The only constant is change: incorporating time-varying network reservations in data centers (2012, methodology): “Hedera:Dynamic.owschedulingfordata center networks.”
Coded MapReduce (2015, methodology): “1) Server 3 sends intermediate pairs (A, 11)[7], (A, 14)[8], (A, 10)[9], (A, 15)[10], (A, 15)[11], (A, 10)[12] and then (B, 21)[7], (B, 21)[8], (B, 21)[9], (B, 21)[10], (B, 21)[11], (B, 21)[12].”
HULA: Scalable Load Balancing Using Programmable Data Planes (2016, background): “However, ECMP suffers from degraded performance [8–12] if two long-running ﬂows are assigned to the same path.”
Software Defined Networking Architecture, Security and Energy Efficiency: A Survey (2017, background): “Hedera has been regarded as an intelligent load balancing scheme capable of handling the large flows [140].”
A roadmap for traffic engineering in SDN-OpenFlow networks (2014, background): “, Hedera [23] and Mahout [33], are proposed.”
c-Through: part-time optics in data centers (2010, methodology): “Hedera: Dynamic .ow scheduling for data center networks.”
DevoFlow: scaling flow management for high-performance networks (2011, methodology): “While OpenFlow was originally proposed for campus and wide-area networks, others have made quantiﬁed arguments that OpenFlow is a viable approach to high-performance networks, such as data center networks [45], and it has been used in proposals for traﬃc management in the data center [5,29].”
AuTO: scaling deep reinforcement learning for datacenter-scale automatic traffic optimization (2018, background): “The main di erence between the two is that SJF schemes [1, 4, 29] require ow size at the start of a ow, while LAS schemes [8, 14, 43] do not.”
ElasticTree: Saving Energy in Data Center Networks (2010, background): “The MST+1 configuration requires one additional edge switch per pod, and one additional switch in the core, to enable any single aggregation or core-level switch to fail without disconnecting a server.”
Augmenting data center networks with multi-gigabit wireless links (2011, background): “This property of real DC workloads renders predictors that rely on identifying elephant flows [2] to be of less use.”
Joint VM placement and routing for data center traffic engineering (2012, background): “A number of recent work [2], [3] proposed the forwarding protocols to make use of path diversity for data center flows.”
Efficient Coflow Scheduling Without Prior Knowledge (2015, background): “Yet, until recently, researchers and practitioners have largely overlooked application-level requirements when improving network-level metrics like flow-level fairness and flow completion time (FCT) [29, 10, 16, 8, 14].”
CONGA: distributed congestion-aware load balancing for datacenters (2015, background): “The source leaf makes load balancing decisions based on per uplink congestion metrics, derived by taking the maximum of the local congestion at the uplink and the remote congestion for the path (or paths) to the destination leaf that originate at the uplink.”
Planck (2014, methodology): “To emulate previous traffic engineering projects that rely on polling [2, 6], we implement a global first fit routing algorithm that greedily reroutes every flow either once a second, which is referenced by the name Poll-1s, or once every 100 ms, which is called Poll-0.”
Network Innovation using OpenFlow: A Survey (2014, methodology): “[83] proposed Hedera, a dynamic flow scheduling method for data center networks.”
Towards predictable datacenter networks (2011, methodology): “Like Hedera [6], we perform centralized rate computation.”
Achieving high utilization with software-driven WAN (2013, background): “Intra-DC traffic management: Many recent works manage intra-DC traffic to better balance load [1, 7, 8] or share among selfish parties [16, 28, 31].”