CSE 5810 Biomedical Informatics and Cloud Computing Zhitong

CSE 5810 Biomedical Informatics and Cloud Computing Zhitong Fei Computer Science & Engineering Department The University of Connecticut CSE 5810: Introduction to Biomedical Informatics Professor: Steven A. Demurjian’s

CSE 5810 Outline 1. Introduction 2 Background and Motivation 3 Big Data for Biomedical Informatics 4 Information Security and Privacy 5 Conclusion

Introduction CSE 5810 m m What is “Cloud Computing”? q A model for enabling ubiquitous, convenient, on-demand access to a shared pool of configurable computing resources. q To provide virtualized IT resources as cloud services by using the Internet technology. Current Situation in Health Care (IOM 2005): q m Limited collaboration, coordination in Health Care. Outcome (IOM 2007): q q q High costs and inefficient patient treatment. Medical errors and increased adverse drug events. Redundancy of clinical data and medical actions. 3

Introduction CSE 5810 4

Background & Motivation m CSE 5810 Why Cloud q Resource outsourcing, utility computing, Large numbers of machines, automated resource management, virtualization and parallel computing. q A style of computing in which dynamically scalable and often virtualized resources are provided as a service over the Internet. q High-Performance Computing(HPC) q Lower cost q Virtualization technology q Parallel characteristics 5

Background & Motivation m CSE 5810 Cloud Computing components q cloud infrastructure, cloud platform and cloud application 6

Background & Motivation m CSE 5810 Classic Cloud architecture q Elastic Compute Cloud (EC 2), Simple Storage Service (S 3), and Simple Queue Service (SQS). 7

Background & Motivation m CSE 5810 Cost q Efficient, Economical, variable cost basis An comparison experiment: Amazon cloud vs Local Compute Cluster 8

Background & Motivation CSE 5810 o o Could not use a 3 -year amortized cost estimate for the cloud-based system. substantial delay required to purchase and install a local cluster was not taken into account. 9

Big Data for Biomedical Informatics CSE 5810 10

Big Data for Biomedical Informatics m CSE 5810 Big Data: Why Bother? q “Big data are data whose scale, diversity, and complexity require new architecture, techniques, algorithms, and analytics to manage it and extract value and hidden knowledge from it. ” 11

Big Data for Biomedical Informatics CSE 5810 m Biomedical Informatics q Genomics, Proteomics, Metabolomics Ø Eg. 1000 Genomes Project, up to 200 terabytes. q q Imaging informatics: at the level of tissues and organs Breakthroughs in diagnosis, prognosis, and highquality healthcare 12

Big Data for Biomedical Informatics CSE 5810 m Cloud Computing and Big Data technologies q Cloud storages allow large-scale q Virtualisation technology -- Virtual machine q A Hypervisor, a virtualisation management layer 13

Big Data for Biomedical Informatics m CSE 5810 Task Scheduling for Cloud computing q Satisfy cloud users and improve profits of cloud providers q Cluster computing: minimize the completion time q Grid computing: improvement of specific performance metrics 14

Information Security and Privacy CSE 5810 15

Information Security and Privacy m Security Risks q Access, Availability, Network load, Integrity, Data Security, Data Location, Data Segregation. q Authenticated users only q Overload and High network load q Natural disasters q Flexibility and scalability q Data integrity m Cloud Security Control q Deterrent Controls, Prevention Controls, Detective Controls, Corrective controls. CSE 5810 16

Information Security and Privacy CSE 5810 17

Information Security and Privacy m CSE 5810 Additional Security in the Cloud q Public encryption key and Private decryption key. q Restricted users accounts q Private clouds, encrypted file systems, and encrypted data volumes 18

Biomedical Informatics and Cloud Computing CSE 5810 19