Name: Prof. Ernesto Damiani

Institute: Center on Cyber-Physical Systems, Khalifa University, Abu Dhabi, UAE Dipartimento di Informatica Giovanni degli Antoni, Università degli Studi di Milano, Milano, Italy

Title: Modeling Artificial Intelligence Big Data Pipelines

Abstract:
In the era of the Internet of Things, huge volumes of highly dimensional data points are made available to applications at an unprecedented velocity. Computing Artificial Intelligence Analytics on such data requires: (i) improving data quality (interpolation, sparsity reduction) to make them suitable for feeding Machine Learning (ML) predictors and classifiers (ii) efficient training and tuning ML predictors and classifiers (iii) operating ML predictors and classifiers in production. In this talk, we discuss how conceptual models can be used to represent key features of such computations, including data ingestion, center-periphery distribution and parallelization, dependencies and potential interferences. Based on such conceptual models, we discuss a methodology and a toolkit for automatic synthesis and deployment of Artificial Intelligence analytics.
Detailed introduction:
Ernesto Damiani is a Professor of computer science at the University of Milan, where he leads the SEcure Service-oriented Architectures Research (SESAR) Lab. Ernesto is the Founding Director of the Center on Cyber-Physical Systems at Khalifa University, in the UAE. Ernesto Damiani received an honorary doctorate from Institut National des Sciences Appliquées de Lyon, France (2017) for his contributions to research and teaching on Big Data analytics. Ernesto is the Principal Investigator of the H2020 TOREADOR project on Big data as a service. His research spans Cyber-security, Big Data and cloud/edge processing, where he has published over 600 peer-reviewed articles and books. He is Distinguished Scientist of ACM and a recipient of the 2017 Stephen Yau Award.

Name: Anqun Pan

Institute: Tencent

Title: Conceptual Modeling On Tencent'€™s Distributed Database Systems

Abstract:
Tencent is the largest Internet service provider in China. Typical services include WeChat, games, payment, cloud storage and computing. Tencent serves billions of users and millions of enterprises, and some services like WeChat, are required to have the ability to handle more than 200,000 TPS requests at peak time. To do this, Tencent has built an elastic and scalable database service system, namely TDSQL, which can efficiently support their ever-growing service requests. TDSQL is deployed and runs on top of more than ten thousands of compute nodes. In this talk, we present the main challenges that we have encountered, and give our practice of conceptual modeling on TDSQL. First, failures of compute nodes often occur in an X86-based large-scale distributed system architecture. To address this issue, we introduce a fault tolerance model to guarantee the high availability of the services. Second, Tencent serves a huge number of requests, while different types of requests require different storage and compute resources. To improve the resource utilization, we propose a resource scheduling model that enables TDSQL to serve the requests elastically. Third, TDSQL provides a hybrid data modeling to support various data models, and develops DBaaS services to serve 10,000 + DB instances. Finally, we present how to fast develop applications in terms of conceptual modeling on top of TDSQL.
Detailed introduction:
Anqun Pan is currently an escalation engineer and the technical director of TEG, Tencent. Tencent is the largest Internet service provider in China, covering social network, finance, entertainment, cloud computing, instant message, tools, AI, etc.. Pan got his bachelor and master degrees from HUST in 2004 and 2007, respectively. He has 10+ years experience in distributed system development, and is leading the team to develop Tencent distributed DBMS, namely TDSQL. TDSQL serves a lot of business and users inside and outside Tencent. The number of digital financial accounts guaranteed by TDSQL is over 28 billion, and the daily trading volume processed by TDSQL exceeds 10 billion.