Caroline P. Lai

The Data Vortex Optical Packet Switched Interconnection Network

A complete review of the data vortex optical packet switched (OPS) interconnection network architecture is presented. The distributed multistage network topology is based on a banyan structure and incorporates a deflection routing scheme ideally suited for implementation with optical components. An implemented 12-port system prototype employs broadband semiconductor optical amplifier switching nodes and is capable of successfully routing multichannel wavelength-division multiplexing packets while maintaining practically error-free signal integrity (BER < 10-12) with median latencies of 110 ns. Packet contentions are resolved without the use of optical buffers via a distributed deflection routing control scheme. The entire payload path in the optical domain exhibits a capacity of nearly 1 Tb/s. Further experimental measurements investigate the OPS interconnection networks flexibility and robustness in terms of optical power dynamic range and network timing. Subsequent experimental investigations support the physical layer scalability of the implemented architecture and serve to substantiate the merits of the data vortex OPS network architectural paradigm. Finally, modified design considerations that aim to increase the network throughput and device-level performance are presented.

Building Data Centers With Optically Connected Memory

Future data centers will require novel, scalable memory architectures capable of sustaining high bandwidths while still achieving low memory access latencies. Electronic interconnects cannot meet the challenges presented by the need for multi-terabit off-chip memory data paths. In this work, the electronic bus between main memory and its host processor is replaced with a circuit-switched optical interconnection network. We investigate the impact of our optically connected memory system on large-scale architectures and experimentally validate the protocol using field-programmable gate array based processor nodes and a custom-designed memory controller. The processor communicates all-optically with multiple synchronous dynamic random access memory nodes using 4 × 2.5-Gb/s wavelength-striped payloads, operating error free with bit-error rates less than 10-12.

Demonstration of 8×40-Gb/s wavelength-striped packet switching in a multi-terabit capacity optical network test-bed

We report on the successful and error-free routing of 8×40-Gb/s wavelength-striped packets in a transparent optical network test-bed, supporting multi-terabit switching capacity. A 0.5-dB power penalty per switch hop is shown at 40 Gb/s.

Broadband Multicasting for Wavelength-Striped Optical Packets

Wavelength-striped optical packet multicasting comprises a potentially important functionality for future energy-efficient network applications. We report on two multicast-capable architectures to experimentally demonstrate multiwavelength packet multicasting in an optical switching fabric testbed. The first design uses programmable packet-splitter-and-delivery that simultaneously supports the nonblocking unicast, multicast, and broadcast of high-bandwidth optical packets with parallel switches. This realization achieves the error-free multicasting of optical messages with 810 Gb/s payloads, with confirmed bit-error rates less than , and scalability of per-channel data rates to 40 Gb/s. We then introduce a second multistage multicasting architecture with lower hardware and energy costs, with the design trade-off of more complex routing logic; the experimental demonstration shows the successful switching and error-free multicasting of 8 10 Gb/s optical packets. The energy costs in terms of the capital and operational expenditures are then compared for the two designs, showing the benefits of the second multicast architecture.

Demonstration of Asynchronous Operation of a Multiwavelength Optical Packet-Switched Fabric

Asynchronous wavelength-striped message routing is experimentally demonstrated for a 4 × 4 optical packet-switched network test-bed. Asynchronous transmission provides increased interconnection network flexibility for future high-performance computing systems. Multiwavelength optical packets with 6 × 10-Gb/s payloads are shown correctly routed asynchronously through the network. Error-free operation with bit-error rates less than 10-12 is confirmed, with an average induced power penalty of 0.5 dB for the six payload wavelengths.

Multichannel 25 Gb/s Low-Power Driver and Transimpedance Amplifier Integrated Circuits for 100 Gb/s Optical Links

Highly integrated electronic driver and receiver ICs with low-power consumption are essential for the development of cost-effective multichannel fiber-optic transceivers with small form factor. This paper presents the latest results of a two-channel 28 Gb/s driver array for optical duobinary modulation and a four-channel 25 Gb/s TIA array suited for both NRZ and optical duobinary detection. This paper demonstrated that 28 Gb/s duobinary signals can be efficiently generated on chip with a delay-and-add digital filter and that the driver power consumption can be significantly reduced by optimizing the drive impedance well above 50 Ω, without degrading the signal quality. To the best of our knowledge, this is the fastest modulator driver with on-chip duobinary encoding and precoding, consuming only 652 mW per channel at a differential output swing of 6 Vpp. The 4 × 25 Gb/s TIA shows a good sensitivity of -10.3 dBm average optical input power at 25 Gb/s for PRBS 231-1 and low power consumption of 77 mW per channel. Both ICs were developed in a 130 nm SiGe BiCMOS process.

Load-Aware Anycast Routing in IP-over-WDM Networks

In this work we propose anycast routing methods to improve the performance of reconfigurable WDM networks under the variations in the IP traffic. We first investigate anycast communication via impairment-aware anycast routing (IAAR); our simulation results show significant improvement in the blocking probability. We also investigate the proposed load-aware anycast routing (LAAR) for the varying traffic model. From the results we observe that LAAR minimizes the lightpath request loss, by dynamically choosing the anycast configuration based on the network load.

Demonstration of Failure Reconfiguration via Cross-Layer Enabled Optical Switching Fabrics

Growing bandwidth requirements of future Internet applications are driving the potential deployment of all-optical packet switching fabrics in next-generation routers. The switching fabric should be capable of executing a fast reconfiguration of its switching state, allowing for the dynamic management of optical packets and the seamless recovery of the fabric, in the case of IP-layer router failures and cross-layer enabled optical-layer signal degradations. We demonstrate a reconfigurable optical switching fabric architecture that uses a field-programmable gate array control plane. Based on the state of a higher-layer router, the switching fabric supports the correct routing and error-free transmission of 10 × 10-Gb/s wavelength-striped optical packets, with bit-error rates less than 10-12 on all payload channels. A power penalty less than 1 dB is shown.

First Demonstration of a Cross-Layer Enabled Network Node

Exploding traffic demands and increasing energy consumptions facing todays networks are driving the designs of next-generation networking technologies. Cross-layer enabled approaches will allow for the packet-level control of the optical layer, to enable dynamic resource allocation and traffic engineering at the physical layer. We demonstrate an intelligent cross-layer enabled network node that can support high-bandwidth, all-optically routed packets, using emerging photonic technologies including optical packet switched fabrics and packet-scale performance monitoring. Using a cross-layer control and management plane, the node can dynamically optimize optical switching based on higher-layer constraints such as quality-of-service and energy consumption, as well as quality-of-transmission metrics such as link integrity and bit-error rates. We demonstrate a first-generation prototype of the cross-layer node, outlining its architecture and major implemented subsystems. The packet-rate physical-layer reconfiguration of the nodes fabric is shown using an implemented performance monitor and control plane. The realized node supports 8 40-Gb/s wavelength-striped optical packets with pseudorandom data with error-free transmission (bit-error rates less than ), in conjunction with the heterogeneous transmission of video traffic using 10-Gigabit Ethernet optical network interface cards based on field-programmable gate arrays.

Cross-layer proactive packet protection switching

To address exponentially-increasing traffic demands, cross-layer optimized architectures will be required to access the optical layer and allow for packet-level control of larger optical flows. In this work, we investigate cross-layer communications that use performance monitoring parameters in a proactive packet protection switching scheme to reroute data traffic. The optical signal quality is captured by bit error ratio measurements and made available to higher-layers, which then trigger flow rerouting based on the per-flow optical signal quality. We experimentally demonstrate cross-layer communications in a test-bed supporting 8 × 10-Gb/s wavelength-striped optical packets. Furthermore, we compare cross-layer packet protection to a fast-reroute scheme in simulation, showing reduced packet loss rates and gains in throughput that are dependent on impairment dynamics and network size.