on self adaptive resource allocation through
play

On Self-adaptive Resource Allocation through Reinforcement Learning - PowerPoint PPT Presentation

Mar 17, 2024 •220 likes •1.33k views

On Self-adaptive Resource Allocation through Reinforcement Learning Jacopo Panerati , Filippo Sironi , Matteo Carminati , Martina Maggio , Giovanni Beltrame , Piotr J. Gmytrasiewicz , Donatella Sciuto and Marco D.

  1. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Machine Learning Methodologies Supervised Learning Classification Algorithms when labels are known to belong to a finite set C Regression Algorithms when labels are known to belong to R Unsupervised Learning Clustering Algorithms when labels are unknown but their cardinality K is assumed be fixed J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 6/31 – mistlab.ca

  2. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Example of Clustering Problem Space Exploration Clustering algorithms can be used to identify patterns in remotely (e.g. in space) sensed data and improve the scientific return by sending to the ground station only statistically significant data [1]. 1 1 http://nssdc.gsfc.nasa.gov/ J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 7/31 – mistlab.ca

  3. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Example of Clustering Problem Space Exploration Clustering algorithms can be used to identify patterns in remotely (e.g. in space) sensed data and improve the scientific return by sending to the ground station only statistically significant data [1]. 1 1 http://nssdc.gsfc.nasa.gov/ J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 7/31 – mistlab.ca

  4. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition. Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  5. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition. Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  6. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition. Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  7. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition. Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  8. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition. Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  9. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition. Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  10. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition . Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  11. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Reinforcements in Behavioural Psychology Definition In behavioural psychology, reinforcement consists of the strengthening of a behaviour associated to a stimulus through its repetition. Pioneers B.F. Skinner (1904-1990), together with E. Thorndike (1874-1949), is considered to be one the fathers of current theories on reinforcement and conditioning [2]. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 8/31 – mistlab.ca

  12. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Pavlov’s Dog A precursor of Skinner theories Ivan Pavlov (1849-1936) made conditioning famous with his experiment of drooling dogs. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 9/31 – mistlab.ca

  13. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Pavlov’s Dog A precursor of Skinner theories Ivan Pavlov (1849-1936) made conditioning famous with his experiment of drooling dogs. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 9/31 – mistlab.ca

  14. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References reinforcement learning in computer science is something a bit different both from supervised/unsupervised learning and reinforcements in behavioural psychology.. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 10/31 – mistlab.ca

  15. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  16. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  17. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  18. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  19. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  20. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  21. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  22. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  23. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  24. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (I) Supervised/Unsupervised Machine Learning data-point → label (or a cluster) Reinforcements in Behavioural Psychology stimulus → behaviour Reinforcement Learning state of the world → action J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 11/31 – mistlab.ca

  25. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (II) Reinforcement Learning state of the world → action → new state of the world → action → .. Because the performance metric of RL (i.e., the collected reward S ) is computed over time, solving a RL problem allows to make • planning • complex, sequential decisions • even counterintuitive decisions J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 12/31 – mistlab.ca

  26. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (II) Reinforcement Learning state of the world → action → new state of the world → action → .. Because the performance metric of RL (i.e., the collected reward S ) is computed over time, solving a RL problem allows to make • planning • complex, sequential decisions • even counterintuitive decisions J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 12/31 – mistlab.ca

  27. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (II) Reinforcement Learning state of the world → action → new state of the world → action → .. Because the performance metric of RL (i.e., the collected reward S ) is computed over time, solving a RL problem allows to make • planning • complex, sequential decisions • even counterintuitive decisions J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 12/31 – mistlab.ca

  28. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (II) Reinforcement Learning state of the world → action → new state of the world → action → .. Because the performance metric of RL (i.e., the collected reward S ) is computed over time, solving a RL problem allows to make • planning • complex, sequential decisions • even counterintuitive decisions J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 12/31 – mistlab.ca

  29. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (II) Reinforcement Learning state of the world → action → new state of the world → action → .. Because the performance metric of RL (i.e., the collected reward S ) is computed over time, solving a RL problem allows to make • planning • complex, sequential decisions • even counterintuitive decisions J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 12/31 – mistlab.ca

  30. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (II) Reinforcement Learning state of the world → action → new state of the world → action → .. Because the performance metric of RL (i.e., the collected reward S ) is computed over time, solving a RL problem allows to make • planning • complex, sequential decisions • even counterintuitive decisions J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 12/31 – mistlab.ca

  31. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (II) Reinforcement Learning state of the world → action → new state of the world → action → .. Because the performance metric of RL (i.e., the collected reward S ) is computed over time, solving a RL problem allows to make • planning • complex, sequential decisions • even counterintuitive decisions J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 12/31 – mistlab.ca

  32. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (III) If today was a sunny day • a classification algorithm would label it as “go to the seaside” • RL would tell you “you might as well study and enjoy the fact that you did not fail your exams later in the summer” RL is not an epicurean carpe diem methodology, but a more farsighted and judicious approach. The point is, not how long you live, but how nobly you live. - Lucius Annaeus Seneca J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 13/31 – mistlab.ca

  33. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (III) If today was a sunny day • a classification algorithm would label it as “go to the seaside” • RL would tell you “you might as well study and enjoy the fact that you did not fail your exams later in the summer” RL is not an epicurean carpe diem methodology, but a more farsighted and judicious approach. The point is, not how long you live, but how nobly you live. - Lucius Annaeus Seneca J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 13/31 – mistlab.ca

  34. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (III) If today was a sunny day • a classification algorithm would label it as “go to the seaside” • RL would tell you “you might as well study and enjoy the fact that you did not fail your exams later in the summer” RL is not an epicurean carpe diem methodology, but a more farsighted and judicious approach. The point is, not how long you live, but how nobly you live. - Lucius Annaeus Seneca J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 13/31 – mistlab.ca

  35. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (III) If today was a sunny day • a classification algorithm would label it as “go to the seaside” • RL would tell you “you might as well study and enjoy the fact that you did not fail your exams later in the summer” RL is not an epicurean carpe diem methodology, but a more farsighted and judicious approach. The point is, not how long you live, but how nobly you live. - Lucius Annaeus Seneca J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 13/31 – mistlab.ca

  36. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (III) If today was a sunny day • a classification algorithm would label it as “go to the seaside” • RL would tell you “you might as well study and enjoy the fact that you did not fail your exams later in the summer” RL is not an epicurean carpe diem methodology, but a more farsighted and judicious approach. The point is, not how long you live, but how nobly you live. - Lucius Annaeus Seneca J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 13/31 – mistlab.ca

  37. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Why Reinforcement Learning is Different (III) If today was a sunny day • a classification algorithm would label it as “go to the seaside” • RL would tell you “you might as well study and enjoy the fact that you did not fail your exams later in the summer” RL is not an epicurean carpe diem methodology, but a more farsighted and judicious approach. The point is, not how long you live, but how nobly you live. - Lucius Annaeus Seneca J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 13/31 – mistlab.ca

  38. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References moving on to self-adaptive computing.. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 14/31 – mistlab.ca

  39. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 15/31 – mistlab.ca

  40. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 15/31 – mistlab.ca

  41. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration Example Multi-platform software Software that is able to run on different hardware configurations seamlessly is a good example of self-configuration. Hardware Detect Config. Run Inst.Tools Software Install J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 16/31 – mistlab.ca

  42. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration Example Multi-platform software Software that is able to run on different hardware configurations seamlessly is a good example of self-configuration. Hardware Detect Config. Run Inst.Tools Software Install J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 16/31 – mistlab.ca

  43. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 17/31 – mistlab.ca

  44. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 17/31 – mistlab.ca

  45. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-optimization Example Smart Video Players Players that can adjust media encoding in order to maintain a certain Quality of Service (QoS) can be considered self-optimizing applications. Video Detect Quality Play Manager Encoder Control J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 18/31 – mistlab.ca

  46. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-optimization Example Smart Video Players Players that can adjust media encoding in order to maintain a certain Quality of Service (QoS) can be considered self-optimizing applications. Video Detect Quality Play Manager Encoder Control J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 18/31 – mistlab.ca

  47. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 19/31 – mistlab.ca

  48. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 19/31 – mistlab.ca

  49. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-healing Example Reconfigurable Logic FPGAs are a good playground for self-healing implementation. Part of the hardware resources can be used to verify the correct functioning of the rest of the logic and force reconfiguration when a fault is detected. Prog.Logic Reconfigure Detect Fault µ Contr. Listener Inform J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 20/31 – mistlab.ca

  50. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-healing Example Reconfigurable Logic FPGAs are a good playground for self-healing implementation. Part of the hardware resources can be used to verify the correct functioning of the rest of the logic and force reconfiguration when a fault is detected. Prog.Logic Reconfigure Detect Fault µ Contr. Listener Inform J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 20/31 – mistlab.ca

  51. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 21/31 – mistlab.ca

  52. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Typical Properties of Self-adaptive Computing Self-configuration The system requires limited or no human intervention in order to set-up. Self-optimization The system is able to achieve user-defined goals autonomously, without human interaction. Self-healing The system can detect and recover from faults without human intervention. Together with self-protection, these are the properties identified in [3] for autonomic system. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 21/31 – mistlab.ca

  53. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Research Question Is RL a suitable approach for self-adaptive computing? J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 22/31 – mistlab.ca

  54. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Case Study Testing Environment • Desktop workstation • Multi-core Intel i7 Processor • Linux-based operating system Objective of our Experiments Enabling self-adaptive properties in applications of the PARSEC[4] benchmark suite through reinforcement learning algorithms. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 23/31 – mistlab.ca

  55. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Case Study Testing Environment • Desktop workstation • Multi-core Intel i7 Processor • Linux-based operating system Objective of our Experiments Enabling self-adaptive properties in applications of the PARSEC[4] benchmark suite through reinforcement learning algorithms. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 23/31 – mistlab.ca

  56. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Case Study Testing Environment • Desktop workstation • Multi-core Intel i7 Processor • Linux-based operating system Objective of our Experiments Enabling self-adaptive properties in applications of the PARSEC[4] benchmark suite through reinforcement learning algorithms. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 23/31 – mistlab.ca

  57. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  58. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  59. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  60. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  61. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  62. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  63. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  64. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Tests Set-Up Reinforcement Learning Framework • A finite set of states S → heart rate of the PARSEC benchmark application measured through Heart Rate Monitor (HRM) APIs [5] • A finite set of actions A → (1) number of cores on which the PARSEC benchmark application is scheduled 2 and (2) CPU frequency 3 • A reward function R ( s ) ∶ S → R → whether a user-defined target (in heartbeats/s) is met or not 2 sched setaffinity system call 3 cpufrequtils package J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 24/31 – mistlab.ca

  65. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  66. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  67. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  68. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  69. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  70. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  71. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  72. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  73. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  74. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  75. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-configuration 1 2 3 4 5 6 7 8 9 10 25 perf. (M options/s) 20 15 10 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 9501000 time(s) blackscholes managed exploiting ADP and core allocation . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 25/31 – mistlab.ca

  76. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-optimization 1 2 3 4 5 6 7 8 9 10 perf. (M exchanges/s) 2 . 5 2 1 . 5 1 0 . 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 time (s) canneal managed exploiting ADP and core allocation. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 26/31 – mistlab.ca

  77. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-optimization 1 2 3 4 5 6 7 8 9 10 perf. (M exchanges/s) 2 . 5 2 1 . 5 1 0 . 5 0 4 3 cores 2 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 time (s) canneal managed exploiting ADP and core allocation. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 26/31 – mistlab.ca

  78. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-healing 1 2 3 4 5 6 7 8 9 10 perf. (M exchanges/s) 2 . 5 2 1 . 5 1 0 . 5 0 4 3 cores 2 1 14 frequency 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 time (s) canneal managed exploiting ADP, core allocation, and frequency scaling. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 27/31 – mistlab.ca

  79. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-healing 1 2 3 4 5 6 7 8 9 10 perf. (M exchanges/s) 2 . 5 2 1 . 5 1 0 . 5 0 4 3 cores 2 1 14 frequency 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 time (s) canneal managed exploiting ADP, core allocation, and frequency scaling . J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 27/31 – mistlab.ca

  80. POLYTECHNIQUE MONTR´ EAL Rationale Reinforcement Learning Self-adaptive Computing Case Study Conclusions References Self-healing 1 2 3 4 5 6 7 8 9 10 perf. (M exchanges/s) 2 . 5 2 1 . 5 1 0 . 5 0 4 3 cores 2 1 14 frequency 1 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 time (s) canneal managed exploiting ADP, core allocation, and frequency scaling. J. Panerati et al. – On Self-adaptive Resource Allocation through Reinforcement Learning 27/31 – mistlab.ca

Recommend

ADEPT Scalability Predictor in Support of Adaptive Resource Allocation IPDPS 2010 Outline

ADEPT Scalability Predictor in Support of Adaptive Resource Allocation IPDPS 2010 Outline

Arash Deshmeh, Jacob Machina, and Angela C. Sodan University of Windsor, Canada ADEPT Scalability Predictor in Support of Adaptive Resource Allocation IPDPS 2010 Outline Background: Adaptive Resource Allocation Related Work Downey

329 views • 22 slides
A Resource Allocation Controller for Cloud-based Adaptive Video Streaming Luca De Cicco , Saverio

A Resource Allocation Controller for Cloud-based Adaptive Video Streaming Luca De Cicco , Saverio

1/16 A Resource Allocation Controller for Cloud-based Adaptive Video Streaming Luca De Cicco , Saverio Mascolo, Dario Calamita Politecnico di Bari, Dipartimento di Ingegneria Elettrica e dellInformazione MCN 2013 - Budapest, Hungary 13 June

564 views • 27 slides
Section 6.1: Resource Allocation Issues Chapter 6: Congestion Control and Resource Allocation

Section 6.1: Resource Allocation Issues Chapter 6: Congestion Control and Resource Allocation

Section 6.1: Resource Allocation Issues Chapter 6: Congestion Control and Resource Allocation CS/ECPE 5516: Comm. Network Prof. Abrams Spring 2000 CS/ECPE5516 1 2 CS/ECPE5516 Whats wrong with doing How to prevent traffic jams

311 views • 6 slides
Resource Allocation Task Force Resource Allocation Task Force Gigi Karmous Edwards Gigi

Resource Allocation Task Force Resource Allocation Task Force Gigi Karmous Edwards Gigi

Resource Allocation Task Force Resource Allocation Task Force Gigi Karmous Edwards Gigi ke@ncsu.edu Gigi_ke@ncsu.edu GLIF Winter Meeting, Hong Kong 2011 Purpose of R Resource Allocation Task Force All i T k F What does the user want

342 views • 7 slides
Resource Allocation and Deadlock Resource Allocation and Deadlock Handling Conditions for

Resource Allocation and Deadlock Resource Allocation and Deadlock Handling Conditions for

Resource Allocation and Deadlock Resource Allocation and Deadlock Handling Conditions for Deadlock [Coffman-etal 1971] 4 conditions must hold simultaneously for a deadlock to occur: deadlock to occur: Mutual exclusion: only one process at

520 views • 31 slides
Dominant Resource Fairness: Fair Allocation of Multiple Resource Types Ali Ghodsi, Matei

Dominant Resource Fairness: Fair Allocation of Multiple Resource Types Ali Ghodsi, Matei

Dominant Resource Fairness: Fair Allocation of Multiple Resource Types Ali Ghodsi, Matei Zaharia, Benjamin Hindman, Andy Konwinski, Scott Shenker, Ion Stoica UC Berkeley Introduction: Resource Allocation There is a need to design fair,

604 views • 22 slides
Adversarial Risk Analysis Models for Urban Security Resource Allocation Urban Security Resource

Adversarial Risk Analysis Models for Urban Security Resource Allocation Urban Security Resource

Adversarial Risk Analysis Models for Urban Security Resource Allocation Urban Security Resource Allocation David Ros Insua, Royal Academy Cesar Gil, U. Rey Juan Carlos Jess Ros, IBM Research YH COST Smart Cities Wshop Paris, September

507 views • 22 slides
D Dominant Resource Fairness (DRF) i t R F i (DRF) Fair Allocation of Multiple Resource Types

D Dominant Resource Fairness (DRF) i t R F i (DRF) Fair Allocation of Multiple Resource Types

D Dominant Resource Fairness (DRF) i t R F i (DRF) Fair Allocation of Multiple Resource Types Ali Ghodsi , Matei Zaharia Benjamin Hindman, Andy Konwinski, B j i Hi d A d K i ki Scott Shenker, Ion Stoica University of California, Berkeley

781 views • 51 slides
Octavian Morariu, Theodor Borangiu, Cristina Morariu, Silviu Raileanu IESS1.6, Bucharest, 2016

Octavian Morariu, Theodor Borangiu, Cristina Morariu, Silviu Raileanu IESS1.6, Bucharest, 2016

Octavian Morariu, Theodor Borangiu, Cristina Morariu, Silviu Raileanu IESS1.6, Bucharest, 2016 Predictive Research capacity provisioning Goals mechanism Queued Adaptive resource resource allocation bus allocation in architecture

436 views • 20 slides
Cost-Effective Resource Allocation for Deploying 1 Vinay Setty vsetty@mpi-inf.mpg.de Pub/Sub

Cost-Effective Resource Allocation for Deploying 1 Vinay Setty vsetty@mpi-inf.mpg.de Pub/Sub

Cost-Effective Resource Allocation for Deploying 1 Vinay Setty vsetty@mpi-inf.mpg.de Pub/Sub on Cloud Published in ICDCS 2014 Roman Vitenberg Maarten van Steen Gunnar Kreitz Guido Urdaneta Cost-Effective Resource Allocation for Deploying

904 views • 41 slides
Heterogeneity-Aware Resource Allocation and Scheduling in the Cloud Gunho Lee (UC Berkeley)

Heterogeneity-Aware Resource Allocation and Scheduling in the Cloud Gunho Lee (UC Berkeley)

Heterogeneity-Aware Resource Allocation and Scheduling in the Cloud Gunho Lee (UC Berkeley) Byung-Gon Chun (Yahoo! Research) Randy H. Katz (UC Berkeley) We have resources and jobs Resource Job/Task Allocate resources (slots) Allocation

755 views • 23 slides
Efficiency and Fairness in Distributed Resource Allocation Ulle Endriss Institute for Logic,

Efficiency and Fairness in Distributed Resource Allocation Ulle Endriss Institute for Logic,

Distributed Resource Allocation TU Delft Agent Colloquium: Nov 2006 Efficiency and Fairness in Distributed Resource Allocation Ulle Endriss Institute for Logic, Language and Computation University of Amsterdam joint work with

405 views • 23 slides
CS 514: Computer Networks Lecture 8: Router-Assisted Resource Allocation Xiaowei Yang

CS 514: Computer Networks Lecture 8: Router-Assisted Resource Allocation Xiaowei Yang

CS 514: Computer Networks Lecture 8: Router-Assisted Resource Allocation Xiaowei Yang xwy@cs.duke.edu Review A fundamental question of networking: who gets to send at what speed? Design Space for resource allocation Router-based

740 views • 46 slides
Gender inequality of educational resource allocation within household in China the effects of

Gender inequality of educational resource allocation within household in China the effects of

Gender inequality of educational resource allocation within household in China the effects of resource-limitation and resource-dilution Yixiao LIU Institute for Population and Development Studies School of Public Policy and Administration

472 views • 21 slides
Chapter 6 Cloud Resource Management and Scheduling Contents Resource management and

Chapter 6 Cloud Resource Management and Scheduling Contents Resource management and

Chapter 6 Cloud Resource Management and Scheduling Contents Resource management and scheduling. Policies and mechanisms. Applications of control theory to cloud resource allocation. Stability of a two-level resource allocation

1.47k views • 39 slides
Resource allocation strategies for multicarrier radio systems Marco Moretti Information

Resource allocation strategies for multicarrier radio systems Marco Moretti Information

Resource allocation strategies for multicarrier radio systems Marco Moretti Information Engineering Department Universit di Pisa marco.moretti@iet.unipi.it Summary Introduction Convex optimization Single-cell resource allocation

1.12k views • 110 slides
Dominant Resource Fairness : Fair Allocation of Multiple Resource Types Ali Ghodsi, Matei

Dominant Resource Fairness : Fair Allocation of Multiple Resource Types Ali Ghodsi, Matei

Dominant Resource Fairness : Fair Allocation of Multiple Resource Types Ali Ghodsi, Matei Zaharia, Benjamin Hindman, Andy Konwinski, Scott Shenker, Ion Stoica University of California, Berkeley Talk by: Yael Amsterdamer Introduction Why

800 views • 22 slides
Dynamic Resource Allocation for Database Servers Running on Virtual Storage Gokul Soundararajan,

Dynamic Resource Allocation for Database Servers Running on Virtual Storage Gokul Soundararajan,

Dynamic Resource Allocation for Database Servers Running on Virtual Storage Gokul Soundararajan, Daniel Lupei, Saeed Ghanbari, Adrian Daniel Popescu, Jin Chen, Cristiana Amza University of Toronto 1 Multi-tier Resource Allocation

922 views • 44 slides
Adaptive combinatorial allocation: How to use limited resources while learning what works

Adaptive combinatorial allocation: How to use limited resources while learning what works

Adaptive combinatorial allocation: How to use limited resources while learning what works Maximilian Kasy Alexander Teytelboym August 2020 Introduction Many policy problems have the following form: Resources, agents, or locations need to

222 views • 20 slides
Crime Research in Geography Resource allocation: we helped design the police Basic Command Unit

Crime Research in Geography Resource allocation: we helped design the police Basic Command Unit

Crime Research in Geography Resource allocation: we helped design the police Basic Command Unit families for the national resource allocation formula in the 1990s. More recently been advising West Yorkshire police. Predictive policing:

415 views • 28 slides
Decentralized Resource Allocation Mechanisms in Networks Tudor Stoenescu Information Science

Decentralized Resource Allocation Mechanisms in Networks Tudor Stoenescu Information Science

Decentralized Resource Allocation Mechanisms in Networks Tudor Stoenescu Information Science and Technology Caltech Organization of the Talk Major issues of resource allocation in networks Overview of fundamental issues in

1.56k views • 89 slides
Criteria for resource allocation ISPC Workshop 9 May 2017 Amsterdam http://ispc.cgiar.org

Criteria for resource allocation ISPC Workshop 9 May 2017 Amsterdam http://ispc.cgiar.org

Criteria for resource allocation ISPC Workshop 9 May 2017 Amsterdam http://ispc.cgiar.org Workshop on Resource Allocation Structure What we did to assess the quality of Phase II CRP proposals Update on work with the System to

449 views • 14 slides
Resource Allocation in Egalitarian Agent Societies Ulle Endriss 1 , Nicolas Maudet 2 , Fariba

Resource Allocation in Egalitarian Agent Societies Ulle Endriss 1 , Nicolas Maudet 2 , Fariba

Resource Allocation in Egalitarian Agent Societies MFI-2003 Resource Allocation in Egalitarian Agent Societies Ulle Endriss 1 , Nicolas Maudet 2 , Fariba Sadri 1 and Francesca Toni 1 1 Department of Computing, Imperial College London Email: {

610 views • 15 slides
Multiagent Resource Allocation: What to optimise, how, and why? Ulle Endriss Imperial College

Multiagent Resource Allocation: What to optimise, how, and why? Ulle Endriss Imperial College

Multiagent Resource Allocation TFG-MARA Ljubljana, 28 February 2005 Multiagent Resource Allocation: What to optimise, how, and why? Ulle Endriss Imperial College London Ulle Endriss, Imperial College London 1 Multiagent Resource Allocation

436 views • 22 slides

More recommend