Read the original HEFT paper: H. Topcuoglu, S. Hariri and M.-Y. Wu, "Performance-effective and low-complexity task scheduling for heterogeneous computing," IEEE Transactions on Parallel and Distributed Systems, 13(3), 2002 (available through the library). If you cannot access it, work from this chapter.
Write a one-page research summary with these headings:
- Problem — what exactly is scheduled, on what platform, with which objective?
- Assumptions — list at least five modelling assumptions and one real-world situation where each might be false.
- Method — the two phases of HEFT and the complexity of the algorithm in terms of e (edges) and p (processors).
- Evaluation — which workloads and metrics were used? What is CCR and why does it matter?
- Limitations — three limitations of HEFT (algorithmic or modelling).
- Your extension — one research idea that addresses a limitation, how you would evaluate it, and which result would convince you it works.