Computer Laboratory

Large-Scale Data Processing and Optimisation (2023-2024 Michaelmas Term)

LSDPO - R244

review_log

Open Source Projects

Reading Club Papers

Mini-Project List

Contact

 

 

 

 

 

 

 

  

 

Overview

This module provides an introduction to large-scale data processing, optimisation, and the impact on computer system's architecture. Large-scale distributed applications with high volume data processing such as training of machine learning will grow ever more in importance. Supporting the design and implementation of robust, secure, and heterogeneous large-scale distributed systems is essential. To deal with distributed systems with a large and complex parameter space, tuning and optimising computer systems is becoming an important and complex task, which also deals with the characteristics of input data and algorithms used in the applications. Algorithm designers are often unaware of the constraints imposed by systems and the best way to consider these when designing algorithms with massive volume of data. On the other hand, computer systems often miss advances in algorithm design that can be used to cut down processing time and scale up systems in terms of the size of the problem they can address. Integrating machine learning approaches (e.g. Bayesian Optimisation, Reinforcement Learning) for system optimisation will also be explored in this course.

Recent computer systems have a massive task to enable heavy data processing for fast training and inference, e.g. dealing with Large Language Model (LLM). This demands more sophisticated system architecture, advanced hardware, and fast tensor operation capable compilers. Complex optimisation plays a crucial role here and it is fundamental at various stages of the development.

On completion of this module, the students should:

  • Understand key concepts of scalable data processing approaches in future computer systems
  • Obtain a clear understanding of building distributed systems using data centric programming and large-scale data processing
  •  Understand a large and complex parameter space in computer system's optimisation and applicability of Machine Learning approach

Module Structure

The module consists of 8 sessions, with 5 sessions on specific aspects of large-scale data processing research. Each session discusses 3-4 papers, led by the assigned students. One session is a hands-on tutorial on MapReduce using data flow programming and/or Deep Neural Networks using Google TensorFlow. The first session advises on how to read/review a paper together with a brief introduction on different perspectives in large-scale data processing and optimisation. The last session is dedicated to the student presentation of open-source project studies. 

Schedule and Reading List

All the sessions will be in the class room (FW26). We will meet every Wednesday (from October 11 to November 29) in 2023. The time slot is 10:00-12:00.1

 2023/10/11 Session 1: Introduction to Large-Scale Data Processing and Optimisation

 2023/10/18 Session 2: Data flow programming

  • Data flow programming and Cluster Computing are essential for a large scale data processing. In ML, dataflow programming holds the key to natural, modular, streamlined ML specification integrated with pre- and post-processing and covering typical ML needs.

1. Yuan Yu, Michael Isard, D. Fetterly, M. Budiu, U. Erlingsson, P.K. Gunda, J. Currey:
DryadLINQ: A System for General-Purpose Distributed Data-Parallel Computing Using a High-Level Language, OSDI, 2008.

2. M. Zaharia, M. Chowdhury, T. Das, A. Dave, J. Ma, M. McCauley, M. Franklin, S. Shenker, I. Stoica:
Resilient Distributed Datasets: A Fault-Tolerant Abstraction for In-Memory Cluster Computing, NSDI, 2013.

3. J. Gjengset, M. Schwarzkopf, J. Behrens, L. T. Araujo, M. Ek, E. Kohler, M. F. Kaashoek and R. Morris:
Noria: dynamic, partially-stateful data-flowfor high-performance web applications, OSDI 2018.

4. J. Dean, S. Ghemawat: MapReduce: Simplified Data Processing on Large Clusters, OSDI, 2004.

Felix Rocke (slides)
5. Derek Murray, Malte Schwarzkopf, Christopher Smowton, Steven Smith, Anil Madhavapeddy and Steven Hand:
Ciel: a universal execution engine for distributed data-flow computing, NSDI 2011. 

6. Naiad Frank McSherry's Talk on Differential Dataflow is here.

6.1. Frank McSherry, Rebecca Isaacs, Michael Isard, and Derek G. Murray,
Composable Incremental and Iterative Data-Parallel Computation with Naiad, no. MSR-TR-2012-105, 2012. 

Daniel Vlasits (slides)
6.2. D. Murray, F. McSherry, R. Isaacs, M. Isard, P. Barham, M. Abadi: Naiad: A Timely Dataflow System, SOSP, 2013. 

6.3. F. McSherry, A. Lattuada, M. Schwarzkopf, T. Roscoe: Shared Arrangements: practical inter-query sharingfor streaming dataflows, VLDB, 2020. 

7. P. Bhatotia, A. Wieder, R. Rodrigues, U. A. Acar, and R. Pasquini: Incoop: MapReduce for incremental computation, ACM SOCC, 2011.

8. M. Abadi et al. Tensorflow: A system for large-scale machine learning. OSDI, 2016.

8.1   M. Abadi et al.: TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems, Preliminary White Paper, 2015.

8.2. M. Abadi, M. Isard and D. Murray: A Computational Model for TensorFlow - An Introduction, MAPL, 2017.

9. M. Looks et al.: Deep Learning with Dynamic Computation Graphs, ICLR, 2017.

Michal Mgeladze-Arciuch (slides)
10. Y. Yu et al.: Dynamic Control Flow in Large-Scale Machine Learning, EuroSys, 2017.

11. R. Nishihara, P. Moritz, et al.: Ray:A Distributed Framework for Emerging AI Applications, OSDI, 2018.

Lauren Wilkes (slides)
12. M. Schaarschmidt, S. Mika, K. Fricke, E. Yoneki: RLgraph: Flexible Computation Graphs for Deep Reinforcement Learning, SysML, 2019.

13. T. Akidau, R. Bradshaw, C. Chambers, S. Chernyak, R.  Fernandez-Moctezuma, R. Lax, S. McVeety, D. Mills, F. Perry, E. Schmidt, S. Whittle: The Dataflow Model: A Practical Approach to Balancing Correctness, Latency, and Cost in Massive-Scale, Unbounded, Out-of-Order Data Processing, VLDB, 2015.

14. S. Li, Y. Zhao, R. Varma, et. al: PyTorch Distributed: Experiences on Accelerating Data Parallel Training, VLDB, 2020.

15. T. Lévai, F. Németh, and G. Rétvári: Batchy Batch scheduling Data Flow Graphs with Service level Objective, NSDI, 2020.

Balázs Tóth (slides)

16 . P. Barham, et al.: Pathways: Asynchronous Distributed Dataflow for ML, MLSys, 2022.  

17 . L. Zheng, et al.: Alpa: Automating Inter- and Intra-Operator Parallelism for Distributed Deep Learning, OSDI, 2022.  

18 . H. Zhu, et al.: MSRL: Distributed Reinforcement Learning with Dataflow Fragments, USENIX_ATC, 2023.  

 2023/10/25 Session 3: Large-scale graph data processing

  • Scalable distributed processing of graph structured data, processing model, and programming model.

Kim Worrall (slides)
1. G. Malewicz, M. Austern, A. Bik, J. Dehnert, I. Horn, N. Leiser, and G. Czajkowski: Pregel: A System for Large-Scale Graph Processing, SIGMOD, 2010.

2. Z. Qian, X. Chen, N. Kang, M. Chen, Y. Yu, T. Moscibroda, Z.Zhang: MadLINQ: large-scale distributed matrix computation for the cloud, EuroSys, 2012.

3. Y. Low,  J. Gonzalez, A. Kyrola, D. Bickson, C. Guestrin, J. Hellerstein: Distributed GraphLab: A Framework for Machine Learning and Data Mining in the Cloud, VLDB, 2012.

Grant Wilkins (slides)
4.J. Gonzalez, Y. Low, H. Gu, D. Bickson, and C. Guestrin: Powergraph: distributed graph-parallel computation on natural graphs. OSDI, 2012.

Stefan Milosevic (slides)
5. J. Shun and G. Blelloch: Ligra: A Lightweight Graph Processing Framework for Shared Memory, PPoPP, 2013. 

6. J.  Gonzalez, R. Xin, A. Dave, D. Crankshaw, M. Franklin, I. Stoica: GraphX: Graph Processing in a Distributed Dataflow Framework, OSDI, 2014. 

7. B. Shao,  H. Wang, Y. Li: Trinity: A Distributed Graph Engine on a Memory Cloud, SIGMOD, 2013.

8. A. Kyrola and G. Blelloch: Graphchi: Large-scale graph computation on just a PC, OSDI, 2012.  

Pranav Talluri (slides)
9. A. Roy, I. Mihailovic, W. Zwaenepoel:   X-Stream: Edge-Centric Graph Processing using Streaming Partitions, SOSP, 2013.

10. A. Roy, L. Bindschaedler, J. Malicevic and W. Zwaenepoel: Chaos: Scale-out Graph Processing from Secondary Storage , SOSP, 2015.

11. F. McSherry, M. Isard and D. Murray: Scalability! But at what COST? , HOTOS, 2015.

12. X. Hu, Y. Tao, C. Chung:  Massive Graph Triangulation, SIGMOD, 2013.

13. W. Xie, G. Wang, D.Bindel, A. Demers, J. Gehrke:  Fast Iterative Graph Computation with Block Updates, VLDB, 2014.

Wenxuan Li (slides)
14. S. Hong, H. Chafi, E. Sedlar, f K.Olukotun: Green-Marl: A DSL for Easy and Efficient Graph Analysis, ASPLOS, 2012.

15. D. Prountzos, R. Manevich, K. Pingali: Elixir: A System for Synthesizing Concurrent Graph Programs, OOPSLA, 2012.

16. D. Nguyen, A. Lenharth, K. Pingali: A Lightweight Infrastructure for Graph Analytics, SOSP 2013.

17. D. Merrill, M. Garland, A. Grimshaw: Scalable GPU Graph Traversal, PPoPP, 2012.

18. A. Gharaibeh, E. Santos-Neto, L. Costa, M. Ripeanu:  Efficient Large-Scale Graph Processing on Hybrid CPU and GPU Systems, IEEE TPC, 2014.

Thomas Yuan (slides)
19. Z. Jia, Y. Kwon, G. Shipman, P. McCormick, M. Erez, A. Aiken: A Distributed Multi-GPU System for Fast Graph Processing, VLDB, 2018.

20. H. Dai, Z. Kozareva, B. Dai, A. Smola and L. Song: Learning Steady-States of Iterative Algorithms over Graphs, ICML, 2018.

21. K. Nilakant, V. Dalibard, A. Roy, and E. Yoneki: PrefEdge: SSD Prefetcher for Large-Scale Graph Traversal.  ACM International Systems and Storage Conference (SYSTOR), 2014.

22. L. Bindschaedler, J. Malicevic, N. Schiper, A. Goel, W. Zwaenepoel: Rock you like a Hurricane: taming skew in large scale anaylitcs. EuroSys, 2018.

 2023/11/01 Session 4: Probabilistic Programming 

  • Role of Probabilistic Programming in computer system's optimisation, Bayesian Optimisation in systems.

Guest lecture: Brooks Paige (UCL, ATI) @11:00.   

Title: Programs as probabilistic models  (slides)

Abstract: Probabilistic models used in quantitative sciences have historically co-evolved with methods for performing inference: specific modeling assumptions are made not because they are appropriate to the application domain, but because they are required to leverage existing software packages or inference methods. The emerging field of probabilistic programming aims to reduce the technical and cognitive overhead for writing and designing novel probabilistic models, by introducing a specialized programming language as an abstraction barrier between modeling and inference. While we would ideally be able to provide “automatic” inference for any probabilistic model, this proves severely challenging for models written in sufficiently expressive languages. In this talk I will discuss some of these difficulties, and provide an introduction and overview of different approaches to probabilistic programming.

Bio: Brooks Paige is an associate professor in machine learning at the University College London AI Centre. He is also a Turing fellow at the Alan Turing Institute, and a statistical ambassador for the Royal Statistical Society. He holds a D.Phil in Engineering Science from the University of Oxford, where he was supervised by Frank Wood; an M.A. in Statistics from Columbia University; and a B.A. in Mathematics from Amherst College.

Reading Club @10:00.

1. E. Bingham et al.: Pyro: Deep Universal Probabilistic Programming, Journal of Machine Learning Research, 2019. 

2. D. Tran et al.: Edward: A library for probabilistic modeling, inference, and criticism, arXiv, 2017.

4. F. Wood, J. van de Meent, V. Mansinghka: A new approach to probabilistic programming inference, AISTATS, 2014.

5. B. Paige and F. Wood: A compilation target for probabilistic programming languages, ICML, 2014.

6. J. Ai et al.: HackPPL: a universal probabilistic programming language, MAPL, 2019.

Jiahao Gai (slides)
7. V. Dalibard, M. Schaarschmidt, and E. Yoneki: BOAT: Building Auto-Tuners with Structured Bayesian Optimization, WWW, 2017. 

8. Ge, H., Xu, K. and Ghahramani, Z.: Turing: A language for flexible probabilistic inference, AISTATS, 2018. 

9. W. Neiswanger et al.: ProBO: Versatile Bayesian Optimization Using Any Probabilistic Programming Language, Arxiv, 2019. 

11. M. Balandat et al.: BOTORCH: Bayesian Optimization in PyTorch, Arxiv 2020. 

12. D. Tran, M. D. Hoffman, D. Moore, C. Suter, S. Vasudevan, A. Radul, M. Johnson, and R. A. Saurous: Simple, Distributed, and Accelerated Probabilistic Programming, NeurIPS, 2018. 

13. G. Baydin et al.: Etalumis: Bringing Probabilistic Programming to Scientific Simulators at Scale, SC, 2018. 

Qianyi Liu (slides)

14 . J. Shao, et al.: Tensor Program Optimization with Probabilistic Programs, NeurIPS, 2022. 

15 . X. Wan, et al.: Bayesian Generational Population-Based Training, ALOE, ICLR, 2022.  

16. Maraval, A. et al.: Sample-Efficient Optimisation with Probabilistic Transformer Surrogates, ArXiv, 2022.

17. J. Snoek, H. Larochelle, and R. Adams: Practical Bayesian Optimization of Machine Learning Algorithms, NIPS, 2012.

18. A. Klein, S. Falkner, S. Bartels, P. Hennig, F. Hutter: Fast Bayesian Optimization of Machine Learning Hyperparameters on Large Datasets, AISTAS, 2017.

19. R. Liaw, E. Liang, R. Nishihara, P. Moritz, J. Gonzalez, I. Stoica: Tune: A Research Platform for Distributed Model Selection and Training, ICML, 2018.  

20. Z. Wang, C. Li, S. Jegelka, and P. Kohli: Batched High-dimensional Bayesian Optimization via Structural Kernel Learning, PLMR, 2017.  

Pedro Sousa (slides)

21. Z. Wang, C. Gehring, P. Kohli, and S. Jegelka: Batched Large-scale Bayesian Optimization in High-dimensional Spaces, AISTATS, 2018.  

22. Grosnit et al.: High-Dimensional Bayesian Optimisation with Variational Autoencoders and Deep Metric Learning, arxiv, 2021.  

23. E. Siivola, J. Gonzalez, A. Paleyes, A. Vehtari: Good practices for Bayesian Optimization of high dimensional structured spaces, arxiv, 2021.  

24. J. Grosse, C. Zhang, and P. Hennig: Probabilistic DAG Search, UAI, 2021.  

25. C. Lin, J. Miano, and E. Dyer: Bayesian optimization for modular black-box systems with switching costs, UAI, 2021.  

26. E. H. Lee, D. Eriksson, V. Perrone and M. Seeger: A Nonmyopic Approach to Cost-Constrained Bayesian Optimization, UAI, 2021.  

 2023/11/08 Session 5: Data Flow Programming Tutorial  

 2023/11/15/ Session 6: Optimisations in ML Compiler

  • ML Compiler, SuperOptimisation, Hi-dimensional Parameter Space, Phase Ordering Problem, Reinforcement Learning.

1. Trofin, M. et al.: MLGO: a Machine Learning Guided Compiler Optimizations Framework, ArXiv, 2021.

2. He, G., Parker, S., Yoneki, E.: X-RLflow: Graph Reinforcement Learning for Neural Network Subgraphs Transformation, MLSys, 2023.

3. Ma, L. et al.: Rammer: Enabling Holistic Deep Learning Compiler Optimizations with rTasks, OSDI, 2020.

Stefan Milosevic (slides)

4. Zheng, L. et al.: EinNet: Optimizing Tensor Programs with Derivation-Based Transformations, OSDI, 2023.

5. Nakandala, S. et al.: A Tensor Compiler for Unified Machine Learning Prediction Serving, OSDI, 2020.

6. Ding, Y. et al.: Hidet: Task-Mapping Programming Paradigm for Deep Learning Tensor Programs, ASPLOS, 2023.

7. T. Chen et al.: Learning to Optimize Tensor Programs, NIPS, 2018.

8. T. Chen, T. Moreau, Z. Jiang, L. Zheng, S. Jiao, E. Yan, H. Shen, M. Cowan, L. Wang, Y. Hu, L. Ceze, C. Guestrin, and A. Krishnamurthy: TVM: An Automated End-to-End Optimizing Compiler for Deep Learning, OSDI, 2018. 

9. T. Chen, T. Moreau, Z. Jiang, L. Zheng, S. Jiao, E. Yan, H. Shen, M. Cowan, L. Wang, Y. Hu, L. Ceze, C. Guestrin, and A. Krishnamurthy: TVM: End-to-End Compilation Stack for Deep Learning, SysML, 2017.

10. Z. Jia, J. Thomas, T. Warszawski, M. Gao, M. Zaharia,  A. Aiken: Optimizing DNN Computation with Relaxed Graph Substitutions, SYSML, 2019.

Pranav Talluri (slides)
11. Z. Jia, O. Padon, J. Thomas, T. Warszawski, M. Zaharia,  A. Aiken: TASO: Optimizing Deep Learning Computation with Automated Generation of Graph Substitutions: SOSP, 2019.

12. KH. Wang, J. Zhai, M. Gao, Z. Ma, S. Tang, L. Zheng, Y. Li, K. Rong, Y. Chen, and Z. Jia: PET: Optimizing Tensor Programs with Partially Equivalent Transformations and Automated Corrections, ODSI, 2021.

13. A. Qiao, S. K. Choe, S. Subramanya, W. Neiswanger, Q. Ho, H. Zhang, G. R. Ganger, E. Xing: Pollux: Co-adaptive Cluster Scheduling for Goodput-Optimized Deep Learning, OSDI, 2021.

Grant Wilkins (slides)
14. Y. Yang, et al.: Equality Saturation for Tensor Graph Superoptimization, MLSys, 2021.  

15. L. Zheng, et al.: TenSet: A Large-scale Program Performance Dataset for Learned Tensor Compilers, NeurIPS, 2021.  

16. C. Nandi, et al.: Rewrite Rule Inference Using Equality Saturation, OOPSLA, 2021.  

17. R. Senanayake, et al.: A sparse iteration space transformation framework for sparse tensor algebra, OOPSLA, 2020.  

Felix Rocke (slides)

18. L. Zheng, et al.: Ansor : Generating High-Performance Tensor Programs for Deep Learning, OSDI, 2020.  

19. M. Li, et al.: AdaTune: Adaptive Tensor Program Compilation Made Efficient, NeurIPS, 2020.  

20. S. Zheng, et al.: FlexTensor: An Automatic Schedule Exploration and Optimization Framework for Tensor Computation on Heterogeneous System, ASPLOS, 2020.  

Thomas Yuan (slides)

21. J. Turner, et al.: Neural Architecture Search as Program Transformation Exploration, ASPLOS, 2021.  

Kim Worrall (slides)

22. Y. Zhou, et al.: Transferable Graph Optimizers for ML Compilers, NEURIPS, 2020.  

23. X. Chen, et al.: Learning to Perform Local Rewriting for Combinatorial Optimization, NeurIPS, 2019.  

24. A. Paliwal et al.: REGAL: Transfer Learning For Fast Optimization of Computation Graphs, arxiv, 2019.  

 2023/11/22 Session 7: Optimisations in Computer Systems 

  • AutoTuner, Automatic Parallelism, Optimisation in Database, Devie Placement/Scheduling.

Balázs Tóth (slides)
1.1 A. Mirhoseini et al.: Device Placement Optimization with Reinforcement Learning, ICML, 2017.
1.2. A. Mirhoseini, A. Goldie, H. Pham, B. Steiner, Q. Le and J. Dean: A Hierarchical Mode for Device Placement, ICLR, 2018.  

2.1 A. Mirhoseini, A. Goldie, et al.: A graph placement methodology for fast chip design, Nature, 2021.  
2.2 A. Mirhoseini, A. Goldie, et al.: Chip Placement with Deep Reinforcement Learning, ISPD, 2020.  

Qianyi Liu (slides)

3. R. Addanki, S. B. Venkatakrishnan, S. Gupta, H. Mao, M. Alizadeh: Placeto: Learning Generalizable Device Placement Algorithms for Distributed Machine Learning, arXiv, 2019. 

4. R. Marcus, P. Negi, Parimarjan, H. Mao, C. Zhang, M. Alizadeh, T. Kraska, O. Papaemmanouil, and N. Tatbul: Neo: A Learned Query Optimizer, VLDB, 2019. 

5. D. Aken et al.: Automatic Database Management System Tuning Through Large-scale Machine Learning, SIGMOD, 2017. 

6. A. Pavlo et al.: Self-Driving Database Management Systems, CIDR, 2017.

7. H. Mao et al.: Park: An Open Platform for Learning-Augmented Computer Systems, OpenReview, 2019.

8. A. Floratou et al.: Dhalion: self-regulating stream processing in Heron, VLDB, 2017.

Jiahao Gai (slides)
9. G. Li, X. Zhou, S. Li, and B. Gao: Qtune: RL for DB query optimisation, VLDB, 2019.

10. D Van Aken, A Pavlo, GJ Gordon, and B Zhang: Automatic database management system tuning through large-scale machine learning, SIGMOD, 2017.

11. A. Pavlo, M. Butrovich, L. Ma, P. Menon, W. Shen Lim, D. Van Aken, W. Zhang: Make Your Database System Dream of Electric Sheep: Towards Self-Driving Operation, VLDB, 2021.

Wenxuan Li (slides)
12. D. Aken , D. Yang, S. Brillard, A. Fiorino, B. Zhang, C. Bilien, and A. Pavlo: An Inquiry into Machine Learning-based Automatic Configuration Tuning Services on Real-World Database Management Systems, VLDB, 2021.

13. Zhao, Y. et al.: TOD: GPU-accelerated Outlier Detection via Tensor Operations, VLDB, 2022.

14. Chowdhery, A. et al.: PaLM: Scaling Language Modeling with Pathways, ArXiv, 2022.

15. E. Liang et al.: RLlib: Abstractions for Distributed Reinforcement Learning, ICML, 2018.

16. Kraska, T., Alizadeh, M., Beutel, A., Chi, E.H., Ding, J., Kristo, A., Leclerc, G., Madden, S., Mao, H. and Nathan, V.: SageDB: A learned database system, CIDR, 2019. 

17. Ma, L., Ding, B., Das, S. and Swaminathan, A.: Active Learning for ML Enhanced Database Systems, SIGMOD, 2020. 

18. A. Kipf, R. Marcus, A. van Renen, M. Stoian, A. Kemper, T. Kraska, and T. Neumann: SOSD: A Benchmark for Learned Indexes, NeurIPS Workshop on ML for Systems, 2019. 

19. D. Ha and J. Schmidhuber: World Models, arXiv, 2018 (https://worldmodels.github.io). 

Lauren Wilkes (slides)
20. R. Marcus, P. Negi, H. Mao, N. Tatbul, M. Alizadeh, and T. Kraska: Bao: Learning to Steer Query Optimizers, VLDB, 2020.

21 . E. Liang, Z. Wu, M. Luo, S. Mika, J. Gonzalez, and I. Stoica: RLlib Flow: Distributed Reinforcement Learning is a Dataflow Problem, NeurIPS, 2021.  

22 . X. Zhang, et al.: Restune: Resource oriented tuning boosted by meta-learning for cloud databases, SIGMOD, 2021.  

23 . j. Ding, et al.: Tsunami: A Learned Multi-dimensional Index for Correlated Data and Skewed Workloads, ArXiv, 2020.  

24 . Ng and S. Russell: Algorithms for inverse reinforcement learning, ICML 2000.  

25. L. Espeholt et al.: IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures, ICML, 2018.

26. T. Li, Z. Xu, J. Tang and Y. Wang: Model-Free Control for Distributed Stream Data Processing using Deep Reinforcement Learning, VLDB, 2018.  

28. J. Ansel et al. Opentuner: an extensible framework for program autotuning. PACT, 2014.

29. J. Ansel et al.: Petabricks: A language and compiler for algorithmic choice. In Proceedings of the 2009 ACM SIGPLAN Conference on Programming Language Design and Implementation, PLDI, 2009.

30. O. Alipourfard et al.: CherryPick: Adaptively Unearthing the Best Cloud Configurations for Big Data Analytics, NSDI, 2017.

31. Z. Jia, M. Zaharia, and A. Aiken: Beyond Data and Model Parallelism for Deep Neural Networks, SYSML, 2019.

32. S. Cereda, S. Valladares, P. Cremonesi and S. Doni: CGPTuner: a Contextual Gaussian Process Bandit Approach for the Automatic Tuning of IT Configurations Under Varying Workload Conditions, VLDB, 2021.

33. Z. Jia, et al.: Unity: Accelerating DNN Training Through Joint Optimization of Algebraic Transformations and Parallelization, OSDI, 2022.  

Pedro Sousa (slides)

34. J. Xing, et al.: Bolt: Bridging the Gap between Auto-tuners and Hardware-native Performance, MLSYS, 2022.  

35. M. Lindauer, et al.: SMAC3: A Versatile Bayesian Optimization Package for Hyperparameter Optimization, Journal of Machine Learning Research, JMLR, 2021.  

  Optimisation in Computer System: Additional Reference Papaers.

  • See the additional papers in optimisation in Computer Systems for extended reading. Here!


 2023/11/29 Session 8: Presentation of Open Source Project Study

      
Presentation of Mini Project (work in progress, plan).

    1. Jiahao Gai (PyTorch)   Adaptive AI for games using DRL algorithms with PyTorch (slides)
    2. Wenxuan Li (LA-MCTS, SMAC)   Enhancing LA-MCTS with SMAC for Heterogeneous Search Space (slides)
    3. Qianyi Liu (LA-MCTS)   Hyperparameter tuning with search space partition: investigating LA-MCTS (slides)
    4. Stefan Milosevic (PyTorch)   Exploring PyTorch by implementing and refine a deep learning model tailored for medical image segmentation (slides)
    5. Felix Rocke (TVM, Triton)   Code-Generation Comparison on TVM and Triton (slides)
    6. Pedro Sousa (World Model)   Leveraging World Models in RL for Circuit Design Optimisation (slides)
    7. Pranav Talluri (Pyro)   Exploring Pyro with deep generative models to investigate Pyro’s suitability as PPL in comparison to other competitors (slides)
    8. Balázs Tóth (Naiad)   Exploring Naiad: a timely dataflow system with X-Stream (slides)
    9. Lauren Wilkes (Pyro)   Causal Inference using Pyro (slides)
    10. Grant Wilkins (PyTorch)   Do We Really Need All of that Scale? (slides)
    11. Kim Worrall (Naiad, Egg)   Parallel distribution of the algorithm over Naiad/Egg graph (slides)
    12. Thomas Yuan (Ray Tune)   Distributed Hyperparameter searching in Ray Tune (slides)

Wrap-up Discussion             

                       

Coursework 1 (Reading Club)

The reading club will require you to read between 1 and 3 papers every week. You need to fill out simple review_log (MS word format, text format) prior to each session and email me by a day before - Tuesday (noon). The minimum requirement of review_log is one per session, but you can read as many as you want and fill the review_log for each paper you read. review_log is not marked but 'tick'.

At each session, 3 - 4 papers are selected under the session topic, and if you are assigned to present your review work, please prepare 15-20 minutes slides for presenting your review work. Your presented material should also be emailed by the following day Thursday. You would present your review work approximately twice during the course. The paper includes following two types and you can focus on the specified aspects upon reviewing the paper.

  1. Full length papers 
    • What is the significant contribution?
    • What is the difference from the existing works?
  2. Short length papers 
    • What is the novel idea?
    • What is required to complete the work?

 Coursework 2 (Reports)

The following three reports are required, which could be extended from the reading assignment of the reading club or a different one within the scope of data centric networking.

  1. Review report on a full length of paper (max 1800 words)
    • Describe the contribution of paper in depth with criticism
    • Crystallise the significant novelty in contrast to the other related work
    • Suggestion for future work
  2. Survey report on sub-topic in data centric networking (~1800 - max 2000 words)
    • Pick up to 5 papers as core papers in your survey scope
    • Read the above and expand your reading through related work
    • Comprehend your view and finish as your survey paper
    • See 'how to write a survey paper' in Assessment section.
  3. Project study and exploration of a prototype (max 2500 words)
    • What is the significance of the project in the research domain?
    • Compare with the similar and succeeding projects
    • Demonstrate the project by exploring its prototype
    • Please email your project selection (MS word format or text format <150 words) by 16:00 on November 10, 2023.
    • Project presentation on November 29, 2023.

The report 1 should be handed in by November 10, 2023 - 12:00 and the report 2 by December 8, 2023 - 12:00 . The report 3 by January 16,  2024 - 12:00 - (Try to finish the mini project by the end of 2023!).

 Assessment

The final grade for the course will be provided as a letter grade or percentage and the assessment will consist of two parts:

  1. 25%: for reading club (Participation,  Presentation + tick of review_log and hands-on tutorial)
    • 10%: Presentation
    • 15%: Participation
  2. 75%: for the three reports
    • 15%: Intensive review report
    • 25%: Survey report
    • 35%: Project study

Open Source Projects

See the candidates of Open Source Projects in data centric networking. The list is not exhausted. If you take anything other than the one in the list, please discuss with me. The purpose of this assignment is to understand the prototype of the proposed architecture, algorithms, and systems through running an actual prototype and present/explain to the other people how the prototype runs, any additional work you have done including your own applications and setup process of the prototype. This experience will give you better understanding of the project. These Open Source Projects come with a set of published papers and you should be able to examine your interests in the paper through running the prototype. Some projects are rather large and may require extensive environment and time; make sure you are able to complete this assignment.

How to Read/Review a Paper

The following papers aid how to read/review a paper.

Further supplement: see ‘how to read/review a paper’ section in Advanced Topics in Computer Systems by Steven Hand.

Presentations

Presentations should be about 15-20 minutes long, where you need to cover the following aspects.

  1. What are the background and the problem domain of the paper? What is the motivation of the presented work? What is the difference from the existing works?  What is the novel idea? How did the paper change/unchange the research in the research community?

  2. What is the significant contribution? How did the authors tackle the problem? Did the authors obtain expected result from their trial?

  3. How do you like the paper and why? What is the takeaway message to you (and to research community)? What is required to complete the work?

The following document aids in presenting a review.

How to write a survey paper

A survey paper provides the readers with an exposition of existing work that is comprehensive and organized. It must expose relevant details associated in the surveying area, but it is important to keep a consistent level of details and to avoid simply listing the different works. Thus a good survey paper should demonstrate a summary of recent research results in a novel way that integrates and adds understanding to work in the field. For example, you can take an approach by classifying the existing literature in your own way; develop a perspective on the area, and evaluate trends. Thus, after defining the scope of your survey, 1) classify and organize the trend, 2) critical evaluation of approaches (pros/cons), and 3) add your analysis or explanation (e.g. table, figure). Also adding reference and pointer to further in-depth information is important (summary from Rich Wolski’s note).

Papers for OS Principles (Distributed Storage and Deterministic Parallelism)

  • Following papers will help you to understand distributed storage and parallelism.
  • Systems Research and System Design
1.  B. Lampson: Hints for Computer Systems Design (Revised), ACM OSR 1983.

  • Distributed Storage
2. S. Ghemawat, H. Gobioff, and S. Leung: The Google File System, ACM SOSP 2003.
3. F. Chang et al: BigTable: A Distributed Storage System for Structured Data, USENIX OSDI 2006.
4. G. DeCandia et al:  Dynamo: Amazon's Highly Available Key-value Store, ACM SOSP 2007.

  • Deterministic Parallelism
5. J. Devietti et al: DMP: Deterministic Shared Memory Multiprocessing, ACM ASPLOS 2009.
6. A. Aviram, et al: Efficient System-Enforced Determistic Parallelism, USENIX OSDI 2010.
7. T. Liu et al: Dthreads: Efficient and Determistic Multithreading, ACM SOSP 2011.

Contact Email

Please email to eiko.yoneki@cl.cam.ac.uk for the question.