Backbone networks operated by cloud providers serving AI training clusters face a growing disconnect between IP-layer capacity planning and optical-layer physical constraints: IP planners decide how much capacity to turn up, but without visibility into spectrum fragmentation, channel placement, ROADM connectivity, and per-band heterogeneity introduced by multiband (C+L) transmission, they produce plans that require additional iterations to converge at the optical layer. Solving the joint IP-optical optimization is computationally prohibitive at production scale. We present FABRIC (Fragmentation-Aware Backbone Resource and IP Capacity planner), a cross-layer spectrum planning system that models spectrum at the lowest frequency-slot granularity, groups IP links by shared-risk link group, computes fragmentation losses from channel data, and feeds these constraints into a weekly capacity optimization cycle. We evaluate FABRIC and show that spectrum-aware planning accurately produces capacity plans grounded in optical-layer reality.

Network design is the process of dimensioning IP capacity over an optical network infrastructure to satisfy a given set of demands and reliability constraints. Specifically, we consider the problem of hose-based cross-layer network design, which seeks to find a minimum cost design that is able to route demand for all hose traffic matrices under all specified failure states. While most network design problems are solved as Mixed Integer Programs, a commercial solver can become intractable due to the scale of today's networks. We demonstrate how the classic Benders decomposition algorithm can be applied and improved for this problem and discuss practical implementation aspects. We showcase a horizontally scalable distributed framework to leverage the decomposable problem structure and solve millions of linear programs in a distributed manner, thereby making the network design problem tractable. In contrast to the conventional approach where failure states and traffic matrices are planned sequentially, the Benders algorithm finds global optimal designs across all traffic matrices and failure states. This leads to network designs with improved solution quality and reliability, with 20--30% less IP capacity and spectrum consumption, 50% less link augments and up to 20x faster runtime that enables design for hyper scale networks in a matter of hours.

Product copywriting is a critical component of e-commerce recommendation platforms. It aims to attract users' interest and improve user experience by highlighting product characteristics with textual descriptions. In this paper, we report our experience deploying the proposed Automatic Product Copywriting Generation (APCG) system into the JD.com e-commerce product recommendation platform. It consists of two main components: 1) natural language generation, which is built from a transformer-pointer network and a pre-trained sequence-to-sequence model based on millions of training data from our in-house platform; and 2) copywriting quality control, which is based on both automatic evaluation and human screening. For selected domains, the models are trained and updated daily with the updated training data. In addition, the model is also used as a real-time writing assistant tool on our live broadcast platform. The APCG system has been deployed in JD.com since Feb 2021. By Sep 2021, it has generated 2.53 million product descriptions, and improved the overall averaged click-through rate (CTR) and the Conversion Rate (CVR) by 4.22% and 3.61%, compared to baselines, respectively on a year-on-year basis. The accumulated Gross Merchandise Volume (GMV) made by our system is improved by 213.42%, compared to the number in Feb 2021.