Yiting Xia

Rethinking the physical layer of data center networks of the next decade: using optics to enable efficient *-cast connectivity

Not only do big data applications impose heavy bandwidth demands, they also have diverse communication patterns denoted as *-cast) that mix together unicast, multicast, incast, and all-to-all-cast. Effectively supporting such traffic demands remains an open problem in data center networking. We propose an unconventional approach that leverages physical layer photonic technologies to build custom communication devices for accelerating each *-cast pattern, and integrates such devices into an application-driven, dynamically configurable photonics accelerated data center network. We present preliminary results from a multicast case study to highlight the potential benefits of this approach.

Blast: Accelerating high-performance data analytics applications by optical multicast

Multicast data dissemination is the performance bottleneck for high-performance data analytics applications in cluster computing, because terabytes of data need to be distributed routinely from a single data source to hundreds of computing servers. The state-of-the-art solutions for delivering these massive data sets all rely on application-layer overlays, which suffer from inherent performance limitations. This paper presents Blast, a system for accelerating data analytics applications by optical multicast. Blast leverages passive optical power splitting to duplicate data at line rate on a physical-layer broadcast medium separate from the packet-switched network core. We implement Blast on a small-scale hardware testbed. Multicast transmission can start 33ms after an application issues the request, resulting in a very small control overhead. We evaluate Blasts performance at the scale of thousands of servers through simulation. Using only a 10Gbps optical uplink per rack, Blast achieves upto 102× better performance than the state-of-the-art solutions even when they are used over a non-blocking core network with a 400Gbps uplink per rack.

A Tale of Two Topologies: Exploring Convertible Data Center Network Architectures with Flat-tree

This paper promotes convertible data center network architectures, which can dynamically change the network topology to combine the benefits of multiple architectures. We propose the flat-tree prototype architecture as the first step to realize this concept. Flat-tree can be implemented as a Clos network and later be converted to approximate random graphs of different sizes, thus achieving both Clos-like implementation simplicity and random-graph-like transmission performance. We present the detailed design for the network architecture and the control system. Simulations using real data center traffic traces show that flat-tree is able to optimize various workloads with different topology options. We implement an example flat-tree network on a 20-switch 24-server testbed. The traffic reaches the maximal throughput in 2.5s after a topology change, proving the feasibility of converting topology at run time. The network core bandwidth is increased by 27.6% just by converting the topology from Clos to approximate random graph. This improvement can be translated into acceleration of applications as we observe reduced communication time in Spark and Hadoop jobs.

Flat-tree: A Convertible Data Center Network Architecture from Clos to Random Graph

Clos networks are easy to implement, whereas random graphs have good performance. We propose flat-tree, a convertible data center network architecture, to combine the best of both worlds. Flat-tree can change the network topology dynamically, so the data center can be implemented as a Clos network and be converted to approximate random graphs of different sizes. To serve the heterogeneous workloads in data centers, flat-tree can organize the network as functionally separate zones each having a different topology. Workloads are placed into suitable zones that best optimize the performance. Simulation results demonstrate that flat-tree has similar performance to random graphs.

An optical programmable network architecture supporting iterative multicast for data-intensive applications

We present an optical programmable network architecture to enable agile and efficient iterative multicasting for cluster computing framework. Support for multiple multicast groups and dynamic group reassignment are experimentally demonstrated.

Stop Rerouting!: Enabling ShareBackup for Failure Recovery in Data Center Networks

This paper introduces sharable backup as a novel solution to failure recovery in data center networks. It allows the entire network to share a small pool of backup devices. This proposal is grounded in three key observations. First, the traditional rerouting-based failure recovery is ineffective, because bandwidth loss from failures degrades application performance drastically. Therefore, failed devices should be replaced to restore bandwidth. Second, failures in data centers are rare but destructive [11], so it is desirable to seek cost-effective backup options. Third, the emergence of configurable data center network architectures promises feasibility of bringing backup devices online dynamically. We design the ShareBackup prototype architecture to realize this idea. Compared to rerouting-based solutions, ShareBackup provides more bandwidth with short path length at low cost.

A cross-layer SDN control plane for optical multicast-featured datacenters

The increasing number of datacenter applications with heavy one-to-many communications has motivated the proposal of optical multicast-featured datacenter architectures. These solutions use optical power splitters to duplicate the optical signal from the input port to all the output ports on the fly, thus are promising for fast, reliable, economical, and energy-efficient group data delivery. Figure 1 illustrates how optical power splitters can be inserted into a datacenter. In case of heavy multicast communications from a particular sender to a group of receivers, an optical power splitter can be connected to these servers statically. They can also be placed to the level of Top-of-Rack (ToR) switches to aggregate traffic, so that a set of servers beneath the sender ToR switch can multicast to another set of servers across the destination ToR switches. Dynamic allocation of optical power splitters can be achieved by having a high-radix optical space switch, e.g. a 3D MEMS switch, as a connectivity substrate [5]. The key challenge of these architectures lies in fitting optical data duplication in the picture of the existing network stack. Optical power splitters can only send data unidirectionally, from input to output but not vice versa. Today’s unicast routing protocols assume bidirectional links and thus are ignorant of these additional optical paths. Since multicast routing protocols depend on unicast routing protocols for topology discovery, multicast trees will not be built over the optical splitters. Utilizing the optical multicast function requires multicast data delivery to be performed over the optical power split-

RPIM: Inferring BGP Routing Policies in ISP Networks

BGP dictates routing between autonomous systems with rich policy mechanisms in todays Internet. Operators translate high-level policy objectives into low-level router configurations without a comprehensive understanding of the actual effects on the network behavior, leaving the routing management an error-prone and time-consuming procedure. A fundamental question is: how to verify the intended routing principles against the actual routing effects of an ISP? In this paper, we develop a Routing Policy Inference Model (RPIM) as the first step towards addressing this fundamental issue. RPIM extracts various policy patterns from the BGP routing tables and translates them into high-level policy objectives of the ISP using a grouping and matching technique. Our work bridges the gap between the high-level policy objectives and the actual routing effects, which provides network operators with a novel approach to verify their policy design principles, thus facilitating the routing management. We evaluate our approach by extensive simulations using the Internet AS-level topology from CAIDA and the real routing data from the Abilene network. Simulation results show that RPIM achieves over 78.94% average inference accuracy in our suggested optimal threshold range. We also verify RPIM on several operating ISPs by the registered policies in an Internet Routing Registry (IRR). A representative case study on AS3292 demonstrates that RPIM effectively infers high-level policy objectives from routing data.

End-to-end demonstration of optical multicasting over a dynamically-reconfigurable hybrid data center network architecture

A hybrid data center network architecture featuring reconfigurable optical functionalities is presented and evaluated on an end-to-end test bed. Physical layer multicast is successfully verified via video streaming and UDP performance measurements.