Search

The correct prediction of subsequential port-to-port routes plays an integral part in maritime logistics and is therefore essential for many further tasks like accurate predictions of the estimated time of arrival. In this paper we present a scalable AI-based approach to predict upcoming port destinations from vessels based on historical AIS data. The presented method is mainly intended as a fill in for cases where the AIS destination entry of a vessel is not interpretable. We describe how one can build a stable and efficient in-database AI solution built on Markov models that are suited for massively parallel prediction tasks with high accuracy. The presented research is part of the PRESEA project (“Real-time based maritime traffic forecast”).

In machine learning, continuously retraining a model guarantees accurate predictions based on the latest data as training input. But to retrieve the latest data from a database, time-consuming extraction is necessary as database systems have rarely been used for operations such as matrix algebra and gradient descent. In this work, we demonstrate that SQL with recursive tables makes it possible to express a complete machine learning pipeline out of data preprocessing, model training and its validation. To facilitate the specification of loss functions, we extend the code-generating database system Umbra by an operator for automatic differentiation for use within recursive tables: With the loss function expressed in SQL as a lambda function, Umbra generates machine code for each partial derivative. We further use automatic differentiation for a dedicated gradient descent operator, which generates LLVM code to train a user-specified model on GPUs. We fine-tune GPU kernels at hardware level to allow a higher throughput and propose non-blocking synchronisation of multiple units. In our evaluation, automatic differentiation accelerated the runtime by the number of cached subexpressions compared to compiling each derivative separately. Our GPU kernels with independent models allowed maximal throughput even for small batch sizes, making machine learning pipelines within SQL more competitive.

Feed composition tables are a commonly used to develop research projects and to develop animal diets. Currently, the National Animal Nutrition Program aims to create a living database containing feed composition information using large datasets provided by commercial laboratories. Using large datasets should ensure representative nutritional values for feeds included in the database; however, managing large datasets requires computer codes to manage and classify feeds correctly. Thus, the objective of this project was to develop 2 models based on supervised machine learning techniques for automated classification of corn grain product samples. The database used in the study contained 88,057 samples of corn grain products resulting from the screening procedure previously described by Tran et al (2016). Two types of supervised machine learning models were developed: decision tree and random forest. Parameters included for feed classification were: dry matter, crude protein, neutral detergent fiber, ash, fat, and starch. Models were trained and validated using 70 and 30% of the dataset, respectively. The decision tree and random forest correctly classified 98.3 and 98.8% of validation dataset, respectively. For each corn grain product the performance of the decision tree and random forest were: corn germ = 91 and 91%; corn germ meal = 97 and 95%; corn gluten feed, dry = 99 and 100%; corn gluten feed, wet = 100 and 100%; corn gluten meal = 99 and 100%; corn grain, dry = 99 and 99%; corn grain, high moisture = 100 and 100%; corn grain, steam-flaked = 34 and 53%; corn hominy feed = 83 and 88%; and corn screenings = 44 and 60%, respectively. In conclusion, the random forest was superior to the decision tree approach for classifying corn grain products. Further development is required to improve the performance of models for classifying corn grain steam-flaked and corn screenings

Oracle Corporation offers two engineered database systems with major innovations for database solutions. In 2008, Oracle Exadata Database Machine was introduced in 2008 for high performance and fully available at an affordable cost with intelligence into storage and networking. In 2011, Oracle introduced Oracle Database Appliance as a family of engineered systems with reduced storage, CPU, and memory resources to fulfill the needs of medium and low-scaled customers.

The Oracle Engineered Systems and other appliances are creating a new subclass of DBAs – the Database Machine Administrator (DMA). The DMA will be working with the hardware, storage, and OS of the machines. This might be viewed similarly to a system administrator, but because it is specific for machines that will house databases and machines that are designed for databases, the knowledge of the databases is valuable. These engineered systems are also not typical standard hardware and other administrators will be looking to the DMA to handle everything from the database, OS, network, storage, and server needs.

Many enterprises have migrated their data from an on-site database to a cloud-based database-as-a-service that handles all database-related administrative tasks while providing a simple SQL interface to the end user. Businesses are also increasingly relying on machine learning to understand their customers and develop new products. Given these converging trends, there is a pressing need for database-as-a-service providers to add support for sophisticated machine learning algorithms to the core functionality of their products.

We demonstrate MLog, a high-level language that integrates machine learning into data management systems. Unlike existing machine learning frameworks (e.g., TensorFlow, Theano, and Caffe), MLog is declarative, in the sense that the system manages all data movement, data persistency, and machine-learning related optimizations (such as data batching) automatically. Our interactive demonstration will show audience how this is achieved based on the novel notion of tensoral views (TViews), which are similar to relational views but operate over tensors with linear algebra. With MLog, users can succinctly specify not only simple models such as SVM (in just two lines), but also sophisticated deep learning models that are not supported by existing in-database analytics systems (e.g., MADlib, PAL, and SciDB), as a series of cascaded TViews. Given the declarative nature of MLog, we further demonstrate how query/program optimization techniques can be leveraged to translate MLog programs into native TensorFlow programs. The performance of the automatically generated Tensor-Flow programs is comparable to that of hand-optimized ones., Proceedings of the VLDB Endowment, 10 (12), ISSN:2150-8097

Database workloads severely underutilize their micro-architectural resources. This results in inefficient use of the chip's limited power budget. While ARM's server-grade processor is a promising energy-efficient alternative, it nevertheless wastes significant portion of the power-budget when running a database workload. In this paper, we systematically explore the ways of improving energy efficiency when running database workloads. We start by describing software-only and hardware-only optimizations, and then discuss two hardware/software co-design approaches. We finally propose combining a diverse set of hardware/software co-designed systems into a heterogeneous, energy-efficient database machine.