Helios: a hybrid electrical/optical switch architecture for modular data centers
By Nathan Farrington, George Porter, Sivasankar Radhakrishnan, Hamid Hajabdolali Bazzaz, Vikram Subramanya, Yeshaiahu Fainman, George C. Papen, Amin Vahdat. Conference on Applications, Technologies, Architectures, and Protocols for Computer Communication, 2010
1,174 citations recorded by the source corpus.
Subfields: Computer Networks and Communications, Electrical and Electronic Engineering
View Paper: https://www.semanticscholar.org/paper/a825470434b4cf7ae528b40616fd979a3b2678c7
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Networks and Communications
- Samir R Das
- Aditya Akella
- Paramvir Victor Bahl
- Vyas Sekar
- Ratul Mahajan
- Nei Kato
- David Wetherall
- Himanshu Gupta
In Electrical and Electronic Engineering
- Ratul Mahajan
- J. Padhye
- Paramvir Victor Bahl
- David Wetherall
- Nei Kato
- Amin Vahdat
- Tat-seng Chua
- Brighten Godfrey
Citations in the corpus
- Planck (2014, methodology): “Workloads We evaluate the traffic engineering schemes on our testbed with a set of synthetic and realistic workloads, similar to previous related work [2, 6, 10].”
- Augmenting data center networks with multi-gigabit wireless links (2011, background): “However, before cost or management considerations come into play, we need to first understand whether 60GHz wireless will perform well in the DC environment despite the challenges of interference and reliability.”
- c-Through: part-time optics in data centers (2010, background): “The second one is parallel work Helios [ 22 ], which explores a similar hybrid electrical/optical data center architecture.”
- FireFly (2014, methodology): “Of course, this is infeasible on several fronts: (1) it requires very high fanout and backplane capacity (potentially nullifying the cost benefits), (2) the cabling complexity would be high [37], and (3) it introduces a single-point of failure [14, 20].”
- TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddings (2023, methodology): “For data centers, there have been many proposals for OCS-based networks [17, 31, 35, 51, 53, 57].”
- Inside the Social Network's (Datacenter) Network (2015, methodology): “If demand can be predicted and/or remains stable over reasonable time periods, it may be feasible to provide circuit-like connectivity between portions of the datacenter [20].”
- Mirror mirror on the ceiling: flexible wireless links for data centers (2012, background): “Optical circuit switching [20, 39, 42] is an alternative for adding burst bandwidth to data centers.”
- Millimeter Wave Communication: A Comprehensive Survey (2018, background): “Consequently, research on network topologies has been carried out to reduce cost and avoid congestion in data centers [164], [165].”
- ProjecToR: Agile Reconfigurable Data Center Interconnect (2016, methodology): “We instrument each server to log application demand, by recording the number of bytes written as part of socket calls.”
- Toward Scalable Systems for Big Data Analytics: A Technology Tutorial (2014, background): “Many optical interconnect schemes [87], [91]–[95] have recently been proposed for data center networks.”
- MicroTE: fine grained traffic engineering for data centers (2011, methodology): “Complementary to our approach, is the use of techniques such as flyways( [12, 19, 8]) that argue for adding extra links or devices as a means of tackling hot spots.”
- A Survey on Optical Interconnects for Data Centers (2012, methodology): “In the evaluation it was shown that for applications in which the traffic demand between some hosts changes slowly, the proposed scheme can reduce significantly the completion time of the applications and offers significantly reduced latency between the nodes connected through the optical circuit. f”
- Trends in big data analytics (2014, background): “As an example of a hybrid electrical/optical switch, Helios [32] is an architecture promising significant reductions in the number of switching elements, cabling, cost, and power consumption.”
- Data Center Networks (2013, neutral): “In this section, we introduce three flexible-topology DCN architectures: c-Through [20], Helios [9] and OSA [5].”
- A Survey on Network Methodologies for Real-Time Analytics of Massive IoT Data and Open Research Issues (2017, background): “Due to the shortcomings of the static architectures for data center networks, many researchers have considered employing the optical switching technology to construct dynamic data center networks [106], [117], [118].”
- Jellyfish: Networking Data Centers Randomly (2011, background): “The end-to-end throughput of a topology depends not only on the capacity of the network, but is also inversely proportional to the amount of network capacity consumed to deliver each byte — that is, the average path length.”