Optical deep learning (DL) accelerators have attracted significant interests due to their latency and power advantages. In this article, we focus on incoherent optical designs. A significant challenge is that there is no known solution to perform single-wavelength accumulation (a key operation required for DL workloads) using incoherent optical signals efficiently. Therefore, we devise a hybrid approach, where accumulation is done in the electrical domain, and multiplication is performed in the optical domain. The key technology enabler of our design is the transistor laser, which performs electrical-to-optical and optical-to-electrical conversions efficiently. Through detailed design and evaluation of our design, along with a comprehensive benchmarking study against state-of-the-art RRAM-based designs, we derive the following key results: (1) For a four-layer multilayer perceptron network, our design achieves 115× and 17.11× improvements in latency and energy, respectively, compared to the RRAM-based design. We can take full advantage of the speed and energy benefits of the optical technology because the inference task can be entirely mapped onto our design. (2) For a complex workload (Resnet50), weight reprogramming is needed, and intermediate results need to be stored/re-fetched to/from memories. In this case, for the same area, our design still outperforms the RRAM-based design by 15.92× in inference latency, and 8.99× in energy.

We present a hybrid optical-electrical analog deep learning (DL) accelerator, the first work to use incoherent optical signals for DL workloads. Incoherent optical designs are more attractive than coherent ones as the former can be more easily realized in practice. However, a significant challenge in analog DL accelerators, where multiply-accumulate operations are dominant, is that there is no known solution to perform accumulation using incoherent optical signals. We overcome this challenge by devising a hybrid approach: accumulation is done in the electrical domain, while multiplication is performed in the optical domain. The key technology enabler of our design is the transistor laser, which performs electrical-to-optical and optical-to-electrical conversions efficiently to tightly integrate electrical and optical devices into compact circuits. As such, our design fully realizes the ultra high-speed and high-energy-efficiency advantages of analog and optical computing. Our evaluation results using the MNIST benchmark show that our design achieves 2214× and 65× improvements in latency and energy, respectively, compared to a state-of-the-art memristor-based analog design.

Sensory IoT (Internet of Things) networks are widely applied and studied in recent years and have demonstrated their unique benefits in various areas. In this paper, we bring the sensor network to an application scenario that has rarely been studied - the academic cleanrooms. We design SENSELET++, a low-cost IoT sensing platform that can collect, manage and analyze a large amount of sensory data from heterogeneous sensors. Furthermore, we design a novel hybrid anomaly detection framework which can detect both time-critical and complex non-critical anomalies. We validate SENSELET++ through the deployment of the sensing platform in a lithography cleanroom. Our results show the scalability, flexibility, and reliability properties of the system design. Also, using real-world sensory data collected by SENSELET++, our system can analyze data streams in real-time and detect shape and trend anomalies with a 91% true positive rate.

We present the first all-optical network, Baldur, to enable power-efficient and high-speed communications in future exascale computing systems. The essence of Baldur is its ability to perform packet routing on-the-fly in the optical domain using an emerging technology called the transistor laser (TL), which presents interesting opportunities and challenges at the system level. Optical packet switching readily eliminates many inefficiencies associated with the crossings between optical and electrical domains. However, TL gates consume high power at the current technology node, which makes TL-based buffering and optical clock recovery impractical. Consequently, we must adopt novel (bufferless and clock-less) architecture and design approaches that are substantially different from those used in current networks. At the architecture level, we support a bufferless design by turning to techniques that have fallen out of favor for current networks. Baldur uses a low-radix, multi-stage network with a simple routing algorithm that drops packets to handle congestion, and we further incorporate path multiplicity and randomness to minimize packet drops. This design also minimizes the number of TL gates needed in each switch. At the logic design level, a non-conventional, length-based data encoding scheme is used to eliminate the need for clock recovery. We thoroughly validate and evaluate Baldur using a circuit simulator and a network simulator. Our results show that Baldur achieves up to 3,000X lower average latency while consuming 3.2X-26.4X less power than various state-of-the art networks under a wide variety of traffic patterns and real workloads, for the scale of 1,024 server nodes. Baldur is also highly scalable, since its power per node stays relatively constant as we increase the network size to over 1 million server nodes, which corresponds to 14.6X-31.0X power improvements compared to state-of-the-art networks at this scale.

We present a new interposer-level optical network based on direct-modulated lasers such as vertical-cavity surface-emitting lasers (VCSELs) or transistor lasers (TLs). Our key observation is that, the physics of these lasers is such that they must transmit significantly more power (21x) than is needed by the receiver. We take advantage of this excess optical power to create a new network architecture called Rome, which splits optical signals using passive splitters to allow flexible bandwidth allocation among different transmitter and receiver pairs while imposing minimal power and design costs. Using multi-chip module GPUs (MCM-GPUs) as a case study, we thoroughly evaluate network power and performance, and show that (1) Rome is capable of efficiently scaling up MCM-GPUs with up to 1024 streaming multiprocessors, and (2) Rome outperforms various competing designs in terms of energy efficiency (by up to 4x) and performance (by up to 143%).

Since the discovery of III-V oxidation by Dallesasse and Holonyak in 1989, significant progress has been made, both technically and commercially, on the use of oxides in compound semiconductor devices. The process-induced modification of refractive index and conductivity allows control of the two carriers of information in optoelectronic systems, the photon and the electron, enabling wide-ranging device applications. Of great technical and commercial importance has been the use of oxidation for the fabrication of high-speed vertical-cavity surface-emitting lasers (VCSELs), first implemented by Deppe's group at The University of Texas at Austin (Austin, TX, USA). Here, the low refractive index III-V oxide's interaction with the optical modes inside the VCSEL creates an optimal overlap of gain and field, enabling lasers with ultralow threshold currents and desirable optical beam properties. The discovery of III-V oxidation, key technical milestones in the fabrication of photonic and electronic devices that use oxidation, and the application to VCSELs are reviewed.