The isoperimetric number of random regular graphs
European Journal of Combinatorics
Uniform generation of random regular graphs of moderate degree
Journal of Algorithms
Introduction to parallel algorithms and architectures: array, trees, hypercubes
Introduction to parallel algorithms and architectures: array, trees, hypercubes
Generalized Hypercube and Hyperbus Structures for a Computer Network
IEEE Transactions on Computers
OpenFlow: enabling innovation in campus networks
ACM SIGCOMM Computer Communication Review
Technology-Driven, Highly-Scalable Dragonfly Topology
ISCA '08 Proceedings of the 35th Annual International Symposium on Computer Architecture
On the second eigenvalue of random regular graphs
SFCS '87 Proceedings of the 28th Annual Symposium on Foundations of Computer Science
A scalable, commodity data center network architecture
Proceedings of the ACM SIGCOMM 2008 conference on Data communication
Dcell: a scalable and fault-tolerant network structure for data centers
Proceedings of the ACM SIGCOMM 2008 conference on Data communication
PortLand: a scalable fault-tolerant layer 2 data center network fabric
Proceedings of the ACM SIGCOMM 2009 conference on Data communication
VL2: a scalable and flexible data center network
Proceedings of the ACM SIGCOMM 2009 conference on Data communication
BCube: a high performance, server-centric network architecture for modular data centers
Proceedings of the ACM SIGCOMM 2009 conference on Data communication
MDCube: a high performance network structure for modular data center interconnection
Proceedings of the 5th international conference on Emerging networking experiments and technologies
c-Through: part-time optics in data centers
Proceedings of the ACM SIGCOMM 2010 conference
Helios: a hybrid electrical/optical switch architecture for modular data centers
Proceedings of the ACM SIGCOMM 2010 conference
SPAIN: COTS data-center Ethernet for multipathing over arbitrary topologies
NSDI'10 Proceedings of the 7th USENIX conference on Networked systems design and implementation
DevoFlow: cost-effective flow management for high performance enterprise networks
Hotnets-IX Proceedings of the 9th ACM SIGCOMM Workshop on Hot Topics in Networks
Proteus: a topology malleable data center network
Hotnets-IX Proceedings of the 9th ACM SIGCOMM Workshop on Hot Topics in Networks
Scafida: a scale-free network inspired data center architecture
ACM SIGCOMM Computer Communication Review
LEGUP: using heterogeneity to reduce the cost of data center network upgrades
Proceedings of the 6th International COnference
A cost comparison of datacenter network architectures
Proceedings of the 6th International COnference
Design, implementation and evaluation of congestion control for multipath TCP
Proceedings of the 8th USENIX conference on Networked systems design and implementation
Proceedings of the 2nd ACM Symposium on Cloud Computing
Camdoop: exploiting in-network aggregation for big data applications
NSDI'12 Proceedings of the 9th USENIX conference on Networked Systems Design and Implementation
Finishing flows quickly with preemptive scheduling
Proceedings of the ACM SIGCOMM 2012 conference on Applications, technologies, architectures, and protocols for computer communication
Finishing flows quickly with preemptive scheduling
ACM SIGCOMM Computer Communication Review - Special october issue SIGCOMM '12
PAST: scalable ethernet for data centers
Proceedings of the 8th international conference on Emerging networking experiments and technologies
Reproducible network experiments using container-based emulation
Proceedings of the 8th international conference on Emerging networking experiments and technologies
Incrementally upgradable data center architecture using hyperbolic tessellations
Computer Networks: The International Journal of Computer and Telecommunications Networking
F10: a fault-tolerant engineered network
nsdi'13 Proceedings of the 10th USENIX conference on Networked Systems Design and Implementation
Proceedings of the ACM International Conference on Computing Frontiers
ElasticSwitch: practical work-conserving bandwidth guarantees for cloud computing
Proceedings of the ACM SIGCOMM 2013 conference on SIGCOMM
A new routing scheme for Jellyfish and its performance with HPC workloads
SC '13 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis
Scaling IP multicast on datacenter topologies
Proceedings of the ninth ACM conference on Emerging networking experiments and technologies
Aspen trees: balancing data center fault tolerance, scalability and cost
Proceedings of the ninth ACM conference on Emerging networking experiments and technologies
Plinko: building provably resilient forwarding tables
Proceedings of the Twelfth ACM Workshop on Hot Topics in Networks
Patch panels in the sky: a case for free-space optics in data centers
Proceedings of the Twelfth ACM Workshop on Hot Topics in Networks
Dahu: commodity switches for direct connect data center networks
ANCS '13 Proceedings of the ninth ACM/IEEE symposium on Architectures for networking and communications systems
Optimal networks from error correcting codes
ANCS '13 Proceedings of the ninth ACM/IEEE symposium on Architectures for networking and communications systems
High throughput data center topology design
NSDI'14 Proceedings of the 11th USENIX Conference on Networked Systems Design and Implementation
| Hi-index | 0.00 |
Industry experience indicates that the ability to incrementally expand data centers is essential. However, existing high-bandwidth network designs have rigid structure that interferes with incremental expansion. We present Jellyfish, a high-capacity network interconnect which, by adopting a random graph topology, yields itself naturally to incremental expansion. Somewhat surprisingly, Jellyfish is more cost-efficient than a fat-tree, supporting as many as 25% more servers at full capacity using the same equipment at the scale of a few thousand nodes, and this advantage improves with scale. Jellyfish also allows great flexibility in building networks with different degrees of oversubscription. However, Jellyfish's unstructured design brings new challenges in routing, physical layout, and wiring. We describe approaches to resolve these challenges, and our evaluation suggests that Jellyfish could be deployed in today's data centers.