Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
parallelism_dinamica [2020/04/07 15:51]
argemiro
parallelism_dinamica [2020/04/08 20:19] (current)
argemiro
Line 3: Line 3:
 ==== What is Parallelism?​ ==== ==== What is Parallelism?​ ====
 \\ \\
-The term Parallelism refers to techniques ​to make programs faster by performing several computations ​at the same time. This requires hardware with multiple processing units. In many cases the sub-computations are of the same structure, but this is not necessary. Graphic computations on a GPU are parallelism. A key problem of parallelism is to reduce data dependencies in order to be able to perform computations on independent computation units with minimal communication between them. To this end, it can even be an advantage to do the same computation twice on different units. +The term Parallelism ​(in Computer Science) ​refers to a technique that allow two or more computing tasks to execute ​at the same time. A key problem of parallelism is to reduce data dependencies in order to be able to perform computations on independent computation units with minimal communication between them. To this end, it can even be an advantage to do the same computation twice on different units.
 \\ \\
 +On modern systems, parallelism can be achieve in two different ways: 
 \\ \\
-===Why is parallelism important?​=== 
- 
-Efficiently using these parallel capabilities is necessary to meet user 
-demands for modelling complex dynamics with large datasets. However, users may be reticent to develop or adapt their current models to take advantage of these parallel environments because such development is often complicated,​ time-consuming,​ and error prone. ​ 
- 
 \\ \\
 +• By executing tasks in the same time slice (and pausing the inactive tasks to give an impression of concurrency);​
 \\ \\
-=== How we change Dinamica EGO infrastructure and make everything as much parallel as possible? === +• By having multiple processing units that physically execute tasks simultaneously.
- +
-Specifically,​ we re-designed the software to leverage available parallel hardware capabilities on any computer, while maintaining flexibility for the user to develop custom models without worrying about such parallelization+
 \\ \\
-{{ :p7.png?800 |}} 
 \\ \\
- +Since modern processor architectures are composed ​of many cores, ​the later method strongly surpasses the former in terms of speed and ease of programmingIn such systemsprogrammers ​must adapt their software to take advantage of Parallelism,​ taking into account concurrent access to dataTo do that, data is usually broken into equal independent slices that are processed in different computing units but fundamental problem arises when this data is not easily separable or is dependent ​of many computing iterationsIn order to deal with this situation, software developers and architects developed many algorithms and techniques to minimize communication between computing units and data dependencyTo this endit can even be an advantage to do the same computation twice on different ​units to prevent synchronization.
-We restructured the core of the software so that models are now broken into tasks according to computer capabilities ​and model semantics (i.e., the rules the computer ​must follow as prescribed by the model)Tasks are then executed by fixed number ​of “workers” using a +
-work-stealing approachThese workers can manipulate ​data in parallel (e.g., parts of an image can be read and written ​ simultaneously) and compute certain tasks using graphics processing ​units. ​+
 \\ \\
 \\ \\
 +====Why is Parallelism important?​====
 \\ \\
 +Efficiently using these parallel capabilities is necessary to meet user demands for modelling complex dynamics with large datasets. However, users may be reticent to develop or adapt their current models to take advantage of these parallel environments because such development is often complicated,​ time-consuming,​ and error prone.
 \\ \\
-===Tasks + Work Stealing in Dinamica EGO 5=== 
- 
-{{ :​task.png?​600 |}} 
-{{ :​apresentacao_do_powerpoint_-_dinamica_ego_5_-_fip_cerrado.pptx_2020-04-07_12-30-32_trim.mp4?​800x800 |}} 
 \\ \\
 +===Graphic computations on a GPU are parallelism!===
 +Besides the general purpose processors, another class of devices like the GPUs are able to offer Parallelism. Such devices can provide a greater level of concurrent operations due to the higher number of processors when compared to traditional CPUs. Since GPU processors are dedicated to certain type of tasks, those devices also need their own set of programming rules and data access.
 \\ \\
 +\\
 +=== How we change Dinamica EGO infrastructure and make everything as much parallel as possible? ===
  
-===Parallelism in Dinamica EGO 4 Vs Parallelism ​in Dinamica EGO 5 === +Before version 5, Parallelism in Dinamica EGO was handled by the Virtual Machine (VM) and the Functor Library (FL) directly, without global coordination. This caused uneven distribution of work between computing systems and sub-optional use of the machine resources. The Task Library was introduced ​in Dinamica EGO 5 to address such problems, by creating a global system for handling and processing work. We restructured the VM and FL sub-systems so that models are now broken into tasks according to computer capabilities and model semantics (i.e., the rules the computer must follow as prescribed by the model). These tasks are distributed to underlying computing systems such as processor cores and graphics processing units (GPU) and the workload is balanced by the use of the work-stealing approach. In version 5, every computing unit is able to manipulate and transform data in parallel (e.g., parts of an image can be read and written simultaneously) and also communicate,​ synchronize and exchange data with its siblings.
-{{ :p1.png?600 |}}+
 \\ \\
-{{ :p2.png?600 |}} 
 \\ \\
-{{ :p3.png?600 |}}+{{ :3c6c66e0-383c-4a1c-a9e6-b9c0d4c09fa4.jpg?800 |}}
 \\ \\
-{{ :p4.png?600 |}} 
 \\ \\
-{{ :p5.png?600 |}}+===Tasks + Work Stealing in Dinamica EGO 5=== 
 +{{ :apresentacao_do_powerpoint_-_dinamica_ego_5_-_fip_cerrado.pptx_2020-04-07_12-30-32_trim.mp4?800x600 ​|}}
 \\ \\
-{{ :p6.png?600 |}}+\\ 
 + 
 +===Parallelism in Dinamica EGO 4 Vs Parallelism in Dinamica EGO 5 === 
 + 
 +\\ 
 +{{:parel_2.png?800 |}}
 \\ \\
 \\ \\
Line 64: Line 60:
 \\ \\
 {{ :​p15.png?​600 |}} {{ :​p15.png?​600 |}}
-\\ 
-{{ :​p16.png?​600 |}} 
 \\ \\
 \\ \\