Comp. Arch. - Lecture 27: VLIW Architectures (Fall 2025)

Comp. Arch. - Lecture 27: VLIW Architectures (Fall 2025)

by Onur Mutlu

Formal & Physical Sciences Computing & Cybersecurity UYComputer scienceUYFComputer architecture and logic design
🎙 Onur Mutlu 👥 64K 📅 January 6, 2026 ⏱ 52 min 👁 932 📄 lecture 🧭 2026-08-15
Available in: English (current) Français

Keywords

VLIWinstruction-level parallelismcompilerstatic schedulinghardware-software co-design

Summary

This lecture, part of the Computer Architecture course at ETH Zürich, introduces VLIW (Very Long Instruction Word) architectures. Professor Onur Mutlu explains the fundamental philosophy of VLIW: shifting the responsibility of finding and scheduling independent instructions from hardware to the compiler. He contrasts this with superscalar execution, where hardware performs dynamic scheduling. The lecture covers the basic concept, the ideal case, and the key challenges, particularly the problem of variable-latency operations such as memory accesses. Mutlu discusses the historical context, including the RISC philosophy and the work of John Cocke, and explains why VLIW has not been as successful as superscalar designs. He also hints at future possibilities with machine learning. The lecture is part of a broader series covering different execution paradigms, with upcoming topics including SIMD, GPUs, and memory-centric computing.

132 words

Critical Evaluation

Value of the Information & Strength of the Argument

The lecture provides a clear and insightful comparison between VLIW and superscalar execution, highlighting the trade-offs between hardware and software complexity. Mutlu’s argumentation is solid, using concrete examples and addressing the fundamental issue of variable-latency operations. He effectively explains why VLIW’s static scheduling is fragile in the presence of unpredictable memory latencies, which is a key reason for its limited adoption. The discussion of energy and power is also valuable, illustrating that low-power hardware may not always lead to lower energy consumption if performance suffers.

Scientific Rigor, Source Quality, Title Accuracy

The lecture is scientifically rigorous, based on established computer architecture principles and research. Mutlu references seminal papers, such as Fisher’s work on VLIW, and provides course materials and recommended readings. The title accurately reflects the content. The lecture is well-structured and technically accurate, with no apparent biases or unsupported claims.

150 words

Title / Content Match

The title accurately reflects the content, which is a lecture on VLIW architectures.

Quality & Reliability

9/10

Lecture by a renowned professor in computer architecture, based on established research and textbook material. The content is well-structured, technically accurate, and includes references to seminal papers and course materials.

Key Moments

Markers derived by PSI from the transcript: the creator did not define chapters.

Cited Sources

Concurring Sources

External References

Contribution & Novelties

The lecture provides a clear and comprehensive overview of VLIW architectures, emphasizing the trade-offs between hardware and software complexity. It highlights the fundamental challenge of variable-latency operations, which is a key reason for VLIW’s limited adoption. The lecture also connects VLIW to broader trends in computer architecture, such as the RISC philosophy and the potential for machine learning to improve compiler scheduling.

Pour aller plus loin :

113 words

Radar Profile

The radar profile shows high scores across all dimensions, indicating a lecture that is rich in information, technically deep, and highly reliable. The balance between quantity and quality is excellent, and the technical level is appropriate for an advanced computer architecture course.

Reliability 9/10