FA26 CS511: Demo Requirements

Overview

Students can earn up to 4% extra credit by presenting a demo related to a lecture topic. Demos should showcase practical implementations, tools, or systems discussed in class.

Requirements

Duration: 20 minutes

Content:

• Demonstrate a working system or tool related to the lecture topic
• Explain the key concepts and how they relate to the course material
• Show interesting features or capabilities
• Discuss practical use cases or applications
• Be prepared to answer questions from the class

Technical Setup:

• Test your demo before class
• Have a backup plan if live demos fail (screenshots, recorded video, etc.)
• Ensure your demo works with the classroom equipment

Sign-up

Use the Demo Sign-up Form to reserve your slot.

Sign-ups open exactly 7 days before each class at 10:00 AM. First come, first served.

Important Concepts that Should be Covered

Lec #	Demo System	Objective
1	MongoDB	Demonstrate document-oriented NoSQL operations including CRUD, indexing, and aggregation pipelines.
2	Dask	Show how parallel data processing works with distributed dataframes and task scheduling.
3	Join through MapReduce	Demonstrate how relational operations, including joins and group-by aggregation, can be performed using the map-reduce interface.
4	HDFS	Explain distributed file system operations, including data get/put and replication.
5	Parquet	Demonstrate columnar storage format features with compression and encoding techniques for analytics.
6	Parquet (JSON)	Show how nested data encoding works for hierarchical structures.
8	PostgreSQL	Demonstrate query optimization by examining execution plans and cost-based query planning.
9	GraphX	Show vertex-centric graph processing model for iterative algorithms like PageRank.
10	Kafka	Demonstrate message streaming with topics, partitions, and consumer groups.
11	Spark - PageRank	Implement iterative graph algorithms with RDD operations and demonstrate convergence criteria.
12	Materialize	Show how incremental view maintenance and streaming SQL provide real-time updates.
13	PostgreSQL MVCC	Demonstrate multi-version concurrency control, transaction isolation levels, and snapshot isolation.
16	RocksDB	Explain key-value store’s basic interface and operations. What happens if the system is killed?
17	Spark - HLL	Demonstrate probabilistic cardinality estimation using HyperLogLog sketches for big data.
21	HNSWlib	Show how hierarchical navigable small world graphs enable approximate nearest neighbor search.
22	FAISS	Demonstrate inverted file index with product quantization for large-scale vector search.
23	e5	Show dense passage retrieval using transformer-based embeddings and similarity search.
25	Llama	Demonstrate language model inference, prompting techniques, and text generation capabilities.
26	mini-swe-agent	Show how AI agents perform automated code generation, testing, and repository navigation.
27	Kishu	Demonstrate process checkpointing, state serialization, and restoration for long-running computations.