Spring 2025: CS 7280 Data Structures & Algorithms for Scalable Computing

Lectures: MoWe / 02:50PM-04:30PM ET at WVH 366

Instructor: Prashant Pandey

• Email: p.pandey [at] northeastern.edu
• Office Hours: MoWe / 01:30PM-02:30PM ET

Teaching Assistant: TBD

• Email:
• Office Hours:

We will use Piazza for all Q&A: TBD

Course Schedule

Course Overview

This course studies advanced data structures and algorithms for handling scalability challenges in large-scale data management and machine learning pipelines. It will cover modern hashing techniques, filters and sketching algorithms, locality-sensitive hashing, succinct data structures, string algorithms, graph algorithms, external memory algorithms, and ML-based learned indexes. This course is appropriate for both undergraduate and graduate students with intermediate data structure and algorithm skills. The course will also require intermediate programming skills in C/C++.

Prerequisites

• For grads: CS 5800 (Grad algorithms)
• For undergrads: CS 3000 (Algorithms and Data)

Please email me if you’re interested in the course but don’t meet the prerequisites.

PhD program: The course fulfills the PhD breath requirements for the “Artificial Intelligence and Data Science” and “Systems and Security " breath area.

Course Topics

• Compact trees
• Succinct data structures
• String algorithms
• Hashing techniques
• Filters and sketches
• Locality sensitive hashing (LSH)
• (Approximate) Nearest neighbor search
• Graph algorithms
• External memory algorithms
• Distributed data structures

Assignments

• Assignments: There will be two assignments, including theoretical problems and/or programming tasks, completed individually.

Projects

• Final project: The main component of the grade is the final group project, where students will:
  • Work in groups of 2-3 (exceptions possible),
  • Implement a project relevant to course material,
  • Ensure the project is unique and requires a significant theoretical or programming effort from all team members. The instructor will provide topic suggestions if needed.

Paper Reading

A set of assigned paper readings is included in the course. Each student is required to pick five papers and submit a one-paragraph synopsis for each. There will be five deadlines throughout the semester. Late submissions will not be accepted without prior approval.

Each review must include:

• What is the problem and why is it hard? (Three sentences)
• An overview of the main idea and contributions (Three sentences)
• How the authors evaluate their solution (Two sentences)

Scribing

• Use this template for scribing.
• Each student may scribe 1-2 lectures depending on class size.
• Send your choice to scribe to the TA on a first-come, first-served basis.
• Submit scribe notes (pdf + source) to the TA.
• Scribe notes are due by 9pm the day after the lecture.

Useful Resources

• The Asymptotic Cheat Sheet for theoretical analyses.
• Assignments, scribe notes, and projects must be typeset in LaTeX. If you’re unfamiliar with LaTeX, refer to this introduction and this Overleaf tutorial.

Grading

• Assignments: 20%
• Final Project: 40%
• Paper Reports: 10%
• Class participation: 20%
• Final Exam: 10%

Late Submission Policy

• No late submissions are allowed. Please plan accordingly.
• For emergencies, prior permission from the instructor is required.

Collaboration and Plagiarism

• Review the Northeastern University Policy on Academic Misconduct.
• You may discuss assignment approaches with others but avoid sharing or copying solutions.
• Declare any collaboration upfront.
• Do not share your code or make it public.
• Using tools like Github Copilot, ChatGPT, or copying code from websites constitutes cheating. Cheating will result in an E in the course and referral to the University Student Behavior Committee.

Note: Any attempt to subvert the grading process is considered cheating. If in doubt, ask.

Schedule

• This course schedule is subject to change through the semester.

• Lecture notes/scribe notes will be uploaded here after each lecture.

• Lectures will not be recorded.