Lecture |
Date |
Topic |
Reading |
Comments |
1 |
Aug 29 |
Overview, Logistics, Goals
(Notes: .ppt, .pdf)
|
|
|
2 |
Aug 31 |
Datacenter Architectures
|
-
The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines
(chapters 1 and 2) [Patrick Wendell,
slides]
-
Above the Clouds: A Berkeley View of Cloud Computing
[Haoyuan Li]
|
|
|
Sep 5 |
Labor Day |
|
|
3 |
Sep 7 |
Cloudera's Software Stack [Invited Lecture: Aaron Myers, Cloudera]
[slides
] |
-
The Datacenter as a Computer: An Introduction to the Design of Warehouse-Scale Machines
(chapters 3, 4, 7)
- Warehouse-Scale Computing: Entering the Teenage Decade (video presentation, Luiz Andre Barroso, Google)
|
|
4 |
Sep 12 |
Technology Trends
[slides
]
|
-
Graphic Processing Units (GPUs):
Computer Architecture, Fifth Edition: A Quantitative Approach, 5th edition
, Chapter 4 (hard copies will be distributed in the class) [Aurojit Panda,
slides]
-
Multicore CPUs: Amdahl's Law in the Multicore Era
[Andrew Wang,
notes]
-
Solid State Devices (SSDs):
PerformanceModeling and Analysis of Flash-based Storage Devices
[Nitesh Mor,
slides]
|
|
5 |
Sep 14 |
Project Suggestions
[slides]
|
The Datacenter Needs an Operating System
|
|
6 |
Sep 19 |
Consistency, Availability, Partitions [slides]
|
-
Cluster-Based Scalable Network Services [Shivaram Venkataraman, slides
]
-
Brewer's Conjecture and the Feasibility of Consistent, Available, Partition-Tolerant Web Services
(
Brewer's CAP Theorem - Julian Browne) [Peter Bailis, slides]
|
|
7 |
Sep 21 |
Paxos
[slides]
|
-
Paxos Made Simple
[Gene Pang, slides]
-
Paxos Made Practical [Gautam Kumar, slides]
-
The Chubby Lock Service for Loosely-Coupled Distributed Systems
[Mosharaf Chowdhury, slides]
|
|
8 |
Sep 26 |
Cluster File Systems
|
-
The Google File System
[Nitesh Mor, slides]
-
Megastore: Providing Scalable, Highly Available Storage for Interactive Services
[Sameer Agarwal, slides]
|
|
9 |
Sep 28 |
Data-flow Computation Frameworks |
-
MapReduce: Simplified Data Processing on Large Clusters
[Angel Rodrigues, slides]
-
Dryad: Distributed Data-Parallel Programs from Sequential Building Blocks
[slides]
|
|
10 |
Oct 3 |
Relational Storage
|
-
HIVE: Data Warehousing & Analytics on Hadoop [Cliff Engle, slides]
-
Pig Latin: A Not-So-Foreign Language for Data Processing [Gene Pang]
-
Scads: Scale-independent storage for social computing applications
[Reynold Xin, slides]
|
|
11 |
Oct 5 |
Column-Oriented Storage Systems [Invited Lecture on HBase: Dhruba Borthakur, Facebook, slides]
|
-
Bigtable: A Distributed Storage System for Structured Data
[Ye Yuan]
-
HBase
|
|
12 |
Oct 10 |
Key-Value Store and Interactive Query Systems |
-
Dynamo: Amazon's Highly Available Key-Value Store [Kay Ousterhout, slides]
-
Dremel: Interactive Analysis of Web-Scale Datasets
[Sameer Agarwal, slides]
|
|
13 |
Oct 12 |
Big Data in the Clouds
|
-
DryadLINQ: A System for General-Purpose Distributed Data-Parallel
Computing Using a High-Level Language
[Tathagata Das, slides]
-
FlumeJava: easy, efficient data-parallel pipelines
[Mosharaf Chowdhury, slides]
|
|
14 |
Oct 17 |
Geographic distributed Storage [Invited Lecture: Raghu Ramakrishnan, Yahoo! Research]
|
-
PNUTS: Yahoo!'s Hosted Data Serving Platform
-
Don't Settle for Eventual:
Scalable Causal Consistency for Wide-Area Storage with COPS
|
|
15 |
Oct 19 |
Programming Languages for the Cloud
|
-
BOOM Analytics: Exploring Data-Centric, Declarative Programming for the Cloud
[Ye Yuan, slides]
-
Erlang - A survey of the language and its industrial applications
[Aurojit Panda, slides]
|
|
16 |
Oct 24 |
DBases in the Cloud
|
-
Relational Cloud: A Database-as-a-Service for the Cloud
[Arka Bhattacharya, slides]
-
Database Scalability, Elasticity, and Autonomy in the
Cloud [Andrew Wang, slides]
|
|
17 |
Oct 26 |
In-Memory Frameworks
|
-
Piccolo: Building Fast, Distributed Programs with Partitioned Tables
[Shivaram Venkataraman]
-
Spark
[Antonio Lupher, slides]
|
|
18 |
Oct 31 |
Multiprgramming for Datacenters [Invited Lecture on Hadoop NextGen: Arun Murthy, Hartonworks]
|
-
Mesos: A Platform for Fine-Grained Resource Sharing in the Data Center
-
Hadoop NextGen
|
|
19 |
Nov 2 |
OSes and Clouds
|
-
An Operating System for Multicore and Clouds: Mechanisms and Implementation
[Gautam Kumar, slides]
-
Akaros
, (a more recent version of the paper is here) [Albert Kim]
|
|
20 |
Nov 7 |
Networking: topologies [Invited Lecture: Amin Vahdat, UCSD/Google]
|
-
VL2: A Scalable and Flexible Data Center Network
-
PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric
-
c-Through: Part-time Optics in Data Centers
|
|
21 |
Nov 9 |
Networking: Traffic Management
|
-
Hedera: Dynamic Flow Scheduling for Data Center Networks [Tathagata Das slides]
-
Managing Data Transfers in Computer Clusters with Orchestra [Justine Sherry, slides]
|
|
22 |
Nov 14 |
Networking: Transport Protocol Improvements [slides]
|
-
Data Center TCP (DCTCP) [Shaddi Hasan, slides]
-
Improving Datacenter Performance and Robustness with Multipath TCP[Anand Iyer, slides]
-
ICTCP: Incast Congestion Control for TCP in Data Center Networks [Hilfi Madari Alkaff, slides]
|
|
23 |
Nov 16 |
Frameworks for Graph Computations [slides]
|
-
Pregel: a system for large-scale graph processing
[Patrick Wendell slides]
-
The GraphLab Abstraction
[Mosharaf Chowdhury, slides]
|
|
24 |
Nov 21 |
Security
|
-
CryptDB: A Practical Encrypted Relational DBMS
[Paul Pearce]
-
Hey, You, Get Off of My Cloud: Exploring Information Leakage in Third-Party Compute Clouds [Edward Wu, slides]
|
25 |
Nov 23 |
Memory Management
|
-
The Case for RAMClouds: Scalable High-Performance Storage Entirely in DRAM
[Sangjin Han, slides]
-
PACMan: Coordinated Memory Caching for Parallel Jobs [Reynold Xin, slides]
|
|
26 |
Nov 28 |
Scheduling and Resource Management [Invited Lecture on resource management challenges at Google: John Wilkes, Google, slides]
|
-
Dominant Resource Fairness (DRF)
-
Modeling and Synthesizing Task Placement Constraints in Google Compute Clusters
|
|
27 |
Nov 30 |
"What is Good Research?"
|
- Hamming's "You and Your Research" talk
- Allen Newell's research style
-
Patterson's "How to Have a Bad Career in Research/Academia" talk
|
No reviews required for this lecture's readings! |
|
Dec 7 |
Poster Session (9:30am-11:30pm, Wozniak Lounge)
|
|
|
|
Dec 9 |
Final project report due (11:59pm)
|
|
|