DidaWiki

Teaching rooms :
Room L1, Polo Fibonacci, first floor.
Room C1, Polo Fibonacci, first floor.
Room C40, ISTI-CNR, Door 19/20, first floor

22/09. Introduction to the course, Logistics. Large-scale problems: status and perspectives.
23/09. Data storage architectures: read-only (search engines), read-write (datastores). Grid computing: elements, definition, architecture, protocols and hourglass model. (slides)
29/09. Grid computing: security, resource management, information management, data management. Architectural models, real-world Grids, Globus Toolkit. (slides)
30/09. Cloud computing: definitions, properties, characteristics. Elasticity, dynamic provisioning and autonomic control. (slides)
03/10. Cloud computing: user benefits, provider benefits, economies of scale. Service models: IAAS, PAAS, SAAS. Deployment models. (slides)
07/10. Cloud computing: comparison with Grids. Cloud programming: fault tolerance, service-oriented architectures, decoupling, autoscaling. (slides)
14/10. Cloud computing: design patterns for Cloud applications. Fault tolerance, decoupling, elasticity implementation, data storage, security. Amazon Web Services overview. (slides)
17/10. Map Reduce: design principles, functional programming, basic programming model, exercises (slides)
21/10. Map Reduce: Hadoop installation, configuration.
24/10. Map Reduce: exercises.
28/10. Map Reduce: combiners, partitioners, scheduling, fault tolerances. HDFS and I/O APIs. Exercises. (slides)
11/11. Map Reduce: exercises.
14/11 Map Reduce: algorithmic patterns. State management, matrices, database operations, graphs. (slides)
18/11 a. Map Reduce: exercises.
18/11 b. Map Reduce: exercises.
21/11 Data models, representation and storage: relational, document, graph models. OLTP vs OLAP. LSM Trees and B-Trees. (slides)
25/11 Data encoding: data flows, backward and forward compatibility, Thrift, Protocol Buffers and Avro encodings. (slides)
28/11 Data management: scalability, performance and fault tolerance of distributed systems. General replicated architecture. Consistency models: strict, linearizable and sequential consistency.(slides)
02/12 Data replication: passive replication, replication log, active partitioning, quorum systems, write conflicts. (slides)
05/12 Sospensione della didattica prot. 57140 del 18/11/2016, Università di Pisa.
09/12 Data replication: client-centric consistency models, FLP and CAP theorems.(slides)
12/12 Time: physical clocks and logical clocks. Happens-before relation and systems of logical clocks. Scalar and vector logical clock systems. (slides)
16/12 Data partitioning: consistent hashing and virtual nodes. (slides)
16/12 Projects discussion.

Grids

Study Notes

1. I. Foster, C. Kesselman, “The Grid 2: Blueprint for a New Computing Infrastructure”, Morgan Kaufmann Publishers Inc., 2003. Chapters 4 and 21.
2. K. Hwang, G. C. Fox, J. Dongarra, “Distributed and Cloud Computing”, Morgan Kaufmann Publishers Inc., 2012. Chapter 7.

Reading Assignments

1. IBM Redbooks, Introduction to Grid Computing, 2003. http://www.redbooks.ibm.com/redbooks/pdfs/sg246778.pdf

Clouds

Study Notes

1. NIST Cloud Computing Reference Architecture
2. Cloud Computing and Grid Computing 360-Degree Compared

Reading Assignments

1. Above the Clouds: A Berkeley View of Cloud Computing
2. Cloud computing and emerging IT platforms: Vision, hype, and reality for delivering computing as the 5th utility

Map Reduce

Study Notes

1. Map-Reduce and the New Software Stack
2. Data-Intensive Text Processing with Map Reduce

Reading Assignments

1. MapReduce: Simplified Data Processing on Large Clusters
2. The Google File System
3. Apapche Hadoop. The scalability update
4. HDFS Scalability: the limits to growth

Virtualization

1. An Overview of Virtual Machine Architectures
2. A Comparison of Software and Hardware Techniques for x86 Virtualization
3. QEMU, a fast and portable dynamic translator

1. The Globus Toolkit Homepage
2. The Hadoop Homepage
3. The Yahoo! Hadoop Tutorial Homepage