Students are expected to:
- present one full conference paper/two short workshop papers during 1-hour
- present individually
- actively participate in the discussions about all the papers.
More specifically
Each presentation will be done by one of the students. The time frame should include ~40-45 minutes for presentation (or 20 minutes for each short paper). Assume people will ask questions during the presentation, and prepare accordingly (i.e., you may need a few spare minutes or slides in case there aren’t many). Each presentation should be followed by a 5-10 minutes discussion. Plan to spend much more than one hour to prepare for presentation and discussion.
Take into consideration the background material required to understand the paper; explaining it might take a while, and can often be as interesting(if not more) than the paper itself.
Remember that slides with lots of text are usually ineffective; Try to aim for sparse text (e.g., 4 bullets of short sentences). Visuals are good; they often help people understand. But visuals (diagrams, graphs,animation) are hard to prepare. Don’t waste too much time on it when preparing the presentation. Always remember: Visuals must be relevant; don’t use them just for entertainment.
Think hard about whether and how to engage the other participants in the discussion. Presentations are more interesting for you and for the audience if everyone is engaged occasionally. People learn more when they are active than when they listen passively. It’s not easy to engage students,but try.
What to Think About When Presenting a Paper
When you are reading a research paper, whether you are presenting it or not, think about the following questions. Being able to answer them shows that you understand the paper. These are also appropriate topics for the discussion.
What is the paper claiming or proposing?
If a system is proposed, how does the proposed system work? details are important. You may need to consult earlier papers, or books, to understand all the details. Consider briefly covering earlier ideas and techniques that must be understood but are not well known. Ideally, listeners would understand the technical ideas in detail.
Are the ideas novel and innovative?
Many good ideas have already been proposed in some context. Try to figure out which ideas and techniques are new, which have been adapted from other contexts, and which are simply tools of the trade.
Did the authors substantiate the claims?
Much of the contribution of papers often depends on analysis of real-world data and/or evaluation and experiments of a proposed system. Thus, much of the effort often lies in producing convincing evidence that the ideas work, and/or that data is reliable and the conclusions sound. Make sure you understand the evidence, not just the ideas.
Do you believe the claims? Do you think that the system works as claimed and or that the conclusions make sense?
Modelling and Accounting:
1. Chasing Carbon: The Elusive Environmental Footprint of Computing, HPCA'20
2. ACT: Designing Sustainable Computer Systems With An Architectural Carbon Modeling Tool, ISCA'22
3. Designing Cloud Servers for Lower Carbon, ISCA’24
4. FOCAL: A First-Order Carbon Model to Assess Processor Sustainability, ASPLOS'24
Recycle/Re-use:
1. Junkyard Computing: Repurposing Discarded Smartphones to Minimize Carbon, ASPLOS'23
2. Transient Internet of Things: Redesigning the Lifetime of Electronics for a More Sustainable Networked Environment, HotCarbon’23 (short)
+
The Environmental Impact of Forever Chemicals in Computing Systems , HotEthics'24 (short)
Datacenters:
1. Yank: Enabling Green Data Centers to Pull the Plug, NSDI'13
2. Carbon Explorer: A Holistic Framework for Designing Carbon Aware Datacenters, ASPLOS'23
3. Sustainable Computing – Without the Hot Air, SIGENERGY Energy Informatics Review'23 (short)
+
Beyond PUE: Flexible Datacenters Empowering the Cloud to Decarbonize, HotCarbon'22 (short)
4. Fairywren: A Sustainable Cache for Emerging Write-Read-Erase Flash Interfaces, OSDI’24
Cloud:
1. On the Limitations of Carbon-Aware Temporal and Spatial Workload Shifting in the Cloud, Eurosys'24
2. Treehouse: A Case For Carbon-Aware Datacenter Software, SigEnergy 2023 (short)
+
Myths and Misconceptions Around Reducing Carbon Embedded in Cloud Platforms, HotCarbon'23 (short)
3. Ecovisor: A Virtual Energy System for Carbon-Efficient Applications, ASPLOS'23
4. Going Green for Less Green: Optimizing the Cost of Reducing Cloud Carbon Emissions, ASPLOS'24
5. Carbon Containers: A System-level Facility for Managing Application-level Carbon Emissions, SoCC'23
AI:
1. Sustainable AI: Environmental Implications, Challenges and Opportunities, MLSys'22
2. The Carbon Footprint of Machine Learning Training Will Plateau, Then Shrink, IEEE Computer’22 (short)
+
Beyond Efficiency: Scaling AI Sustainably, IEEE MICRO’24 (short)
3. Making AI Less "Thirsty": Uncovering and Addressing the Secret Water Footprint of AI Models, CACM’24
HPC:
1. GreenHadoop: Leveraging Green Energy in Data-Processing Frameworks, Eurosys'12
2. Toward Sustainable HPC: Carbon Footprint Estimation and Environmental Implications of HPC Systems, SC'23
Web:
1. Quantifying Carbon Emissions due to Online Third-Party Tracking, arxiv 2023
2. Understanding and Mitigating Webpage Data Bloat: Causes and Preventive Measures, HotCarbon’24 (short)
+
Streaming the future of sustainability, HotCarbon’24 (short)