The ground-state conformation of a molecule is often decisive for its properties. However, experimental or computational methods, such as density functional theory (DFT), are time-consuming and labor-intensive for obtaining this conformation. Deep learning (DL) based molecular representation learning (MRL) has made significant advancements in molecular modeling and has achieved remarkable results in various tasks. Consequently, it has emerged as a promising approach for directly predicting the ground-state conformation of molecules. In this regard, we introduce GTMGC, a novel network based on Graph-Transformer (GT) that seamlessly predicts the spatial configuration of molecules in a 3D space from their 2D topological architecture in an end-to-end manner. Moreover, we propose a novel self-attention mechanism called Molecule Structural Residual Self-Attention (MSRSA) for molecular structure modeling. This mechanism not only guarantees high model performance and easy implementation but also lends itself well to other molecular modeling tasks. Our method has been evaluated on the Molecule3D benchmark dataset and the QM9 dataset. Experimental results demonstrate that our approach achieves remarkable performance and outperforms current state-of-the-art methods as well as the widely used open-source software RDkit.

Napa holds Google's critical data warehouses in log-structured merge trees for real-time data ingestion and sub-second response for billions of queries per day. These queries are often multi-key look-ups in highly skewed tables and indexes. In our production experience, only progressive query-specific partitioning can achieve Napa's strict query latency SLOs. Here we advocate good-enough partitioning that keeps the per-query partitioning time low without risking uneven work distribution. Our design combines pragmatic system choices and algorithmic innovations. For instance, B-trees are augmented with statistics of key distributions, thus serving the dual purpose of aiding lookups and partitioning. Furthermore, progressive partitioning is designed to be "good enough" thereby balancing partitioning time with performance. The resulting system is robust and successfully serves day-in-day-out billions of queries with very high quality of service forming a core infrastructure at Google.

Large-scale data transport for data-intensive sciences is a complex multidimensional challenge. The challenge includes optimizing the end-to-end Big Data movement performance in real-time, supporting direct remote data access using NVMe over Fabrics (NVMeoF) and deploying to existing research platforms. AIDTN is the first effort to provide a unique AI system designed to incorporate NVMe over Fabrics (NVMeoF) and optimize coordination among multiple components supporting large-scale, multi-domain Wide Area Network (WAN) data-intensive science. AIDTN's research objective is to integrate next-generation storage architecture using NVMeoF, specialized network design using high-performance network appliances, Data Transfer Nodes (DTNs), catalysts in driving data transport, and a unique AI system explicitly designed for high-performance data movement challenges. AIDTN is the first system that uses network and system features to predict the end-to-end performance of high-performance data movement and further extends the model with NVMe-specific features for NVMeoF remote data access. As a result, AIDTN improves data movement performance by up to 284% while minimizing packet loss compared to other heuristics approaches. It also has a prediction error rate as low as 0.16 compared to AI models with the only network (error rate = 0.29) or network and system features (error rate = 0.19).

Google services continuously generate vast amounts of application data. This data provides valuable insights to business users. We need to store and serve these planet-scale data sets under the extremely demanding requirements of scalability, sub-second query response times, availability, and strong consistency; all this while ingesting a massive stream of updates from applications used around the globe. We have developed and deployed in production an analytical data management system, Napa, to meet these requirements. Napa is the backend for numerous clients in Google. These clients have a strong expectation of variance-free, robust query performance. At its core, Napa's principal technologies for robust query performance include the aggressive use of materialized views, which are maintained consistently as new data is ingested across multiple data centers. Our clients also demand flexibility in being able to adjust their query performance, data freshness, and costs to suit their unique needs. Robust query processing and flexible configuration of client databases are the hallmark of Napa design. Most of the related work in this area takes advantage of full flexibility to design the whole system without the need to support a diverse set of preexisting use cases. In comparison, a particular challenge we faced is that Napa needs to deal with hard constraints from existing applications and infrastructure, so we could not do a "green field" system, but rather had to satisfy existing constraints. These constraints led us to make particular design decisions and also devise new techniques to meet the challenges. In this paper, we share our experiences in designing, implementing, deploying, and running Napa in production with some of Google's most demanding applications.

This paper presents a new matching-based framework for semi-supervised video object segmentation (VOS). Recently, state-of-the-art VOS performance has been achieved by matching-based algorithms, in which feature banks are created to store features for region matching and classification. However, how to effectively organize information in the continuously growing feature bank remains under-explored, and this leads to an inefficient design of the bank. We introduced an adaptive feature bank update scheme to dynamically absorb new features and discard obsolete features. We also designed a new confidence loss and a fine-grained segmentation module to enhance the segmentation accuracy in uncertain regions. On public benchmarks, our algorithm outperforms existing state-of-the-arts.

New compression innovations and high-density silicon technology enable us to introduce main-memory compression. This technology is able to achieve, in most cases, 2:1 or better compression without impacting performance. It provides an enormous cost/performance advantage, given the cost content of memory in modern enterprise servers. The complex and highly parallel data manipulations central to this compression implementation would, if unprotected by extensive error detection and error correction techniques, offer several potential data integrity exposures. This paper describes the memory subsystem of an enterprise class server with a compressed mainstore and the methods which have been employed to guarantee the integrity of the compressed data. These methods consist of a novel ECC algorithm which includes address information in the code words, the use of CRC codes for compressed data blocks, and various consistency checks on the memory management structures used in the management of a compressed mainstore.