SCS Faculty Candidate
Livestream - ET - from Newell-Simon Hall 4305 (Research Talk)
ELBA GARZA , Ph.D. CandidateDepartment of Computer Science and Computer EngineeringTexas A&M University
Developing Robust Microarchitectural Predictive Structures for the Front End
Research Talk: Developing Robust Microarchitectural Predictive Structures for the Front End
Modern processors rely on predictive structures and policies for making important control and data flow execution decisions. To maintain forward progress while awaiting resolving instructions, early computer architects looked to speculation–effectively acting on predicted directions of control flow. Without quick yet accurate prediction, processor architectures may still be relegated to the age of stall-heavy and heavily sequential execution. In this talk, I outline my research contributions toward making microarchitectural predictive structures and policies more accurate and resilient to changing computing demands. Structures and policies outlined include those for the branch target buffer, instruction cache, and translation lookaside buffer.
- Teaching Demo: 19 January 2022 - 10:00 am ET
- Faculty Candidate Research Talk: 20 January 2021 - 10:00 am ET
Teaching Demo: Principles of Caching within the Context of Computer Hardware and Organization
In this mock lecture, I present the principles of caching within the context of a computer hardware and organization course. This mock lecture presents the first 45 minutes of a typical 75-minute lecture for a semester-long course in computer organization, or the equivalent of 15-213. This lecture assumes students’ previous exposure to the lecture’s topics via a pre-course Computer Organization and Design (6th edition, MIPS) ZyBook interactive activity; however, it is not necessary to complete or purchase for the mock lecture. Concepts are presented using in-classroom activities, and abstraction via real-life metaphors for fundamental understanding.
Elba Garza is a final-year Computer Science PhD student at Texas A&M University working under Daniel A. Jiménez. Her work focuses on making hardware predictive structures & policies (e.g., branch prediction, prefetching, cache replacement) more resilient to evolving computing demands. She holds a BSc in Computer Science from Columbia University, and an MSc in the same from Princeton University. Her work has been featured in top-tier computer architecture conferences including ISCA and MICRO. Her research has earned accolades, including the 2020 IEEE CS Lance Stafford Larson Student Award. She was named a 2020 Rising Star in EECS and also recently named a 2021 Google Scholar. After returning to graduate school for her PhD and seeing fellow graduate students struggle with mental health issues arising from the general toll of academic life, she helped co-found the Computer Architecture Student Association, or CASA. CASA is an independent student-run organization with the express purpose of developing and fostering a positive and inviting student community within computer architecture.
Faculty Host: Brian Railing
Computer Science Department
→ Presentation in Newell-Simon 4305.
→ Due to modified campus posture, the talks will be livestreamed. See announcement.
For More Information: