| 1. | The max degree option on dsntip4 determines the maximum degree of parallelism maximum number of parallel tasks
Dsntip4上的max degree选项决定了最大并行度（并行任务的最大数量） 。 |
| 2. | I recommend that you evaluate the virtual storage capacity and constraints in your z os environment , and adjust this parameter as needed , so that db2 will not create more parallel tasks than your virtual storage can handle
我建议您先估计z / os环境中的虚拟存储能力和局限性，这样db2就不至于创建多于虚拟存储所能处理的并行任务。 |
| 3. | For example , an application consisting of many independent parallel tasks executes much faster when the tasks are spread across a large number of compute resources than by using a single server or cluster
例如，与使用单个服务器或群集执行的执行速度相比，包含多个独立并行任务的应用程序在将任务划分到大量计算资源上，其执行速度要更快一些。 |
| 4. | Some improvements are made to enhance the generic algorithm ' s stability self adaptability and optimization efficiency . based on the ttg model , the co - synthesis method can map parallel task graphs to multiple processing unlts
本文提出了基于改进遗传算法解决软硬件协同综合问题的新方法，基于ttg模型，该方法能够实现多并发任务流在多处理单元结构上的映射。 |
| 5. | As for the sequential and parallel processing , which exists in the field of cscw , temporal constrains are used to describe the temporal relations between the sequential and parallel tasks . an application , which contains sequential and parallel sub - tasks , is shown as an example at last
同时，文中分析了协同工作中存在的串行和并行工作，采用时态约束描述了串、并行任务间的时序关系，并描述了一个并行协同工作的实例。 |
| 6. | Although this and my previous twelve scheduling tips are not an exhaustive list of ways to improve scheduling , i m confident that acknowledging and plotting your parallel tasks in this way will provide some ease to organizations that are taking on increasing workload
尽管这个和我以前的12点时间安排的提示都不是对改进时间安排的详尽的方法列表，我相信用这种方法来承认和测绘并发的任务可以为那些负担增加工作负荷的组织提供一些方便。 |
| 7. | Firstly , we need consider more in cooperation and message - pass between parallel tasks . secondly , the parallel design pattern is different from that of sing - alone program . thirdly , parallel program has to call complicated apis to accomplish data transfer
首先，并行程序的分析与设计需要更多的考虑各并行任务的协同和消息通信机制；其次，并行程序的编写模式与单机程序不同；最后，并行程序需要调用并行平台的复杂的通信原语来完成数据的传送。 |
| 8. | In chapter 3 we construct two approximation algorithms which applying bin packing algorithms for scheduling problems , one is ff ( first fit ) algorithm used in parallel machine scheduling problem pm / / dj = d / n which has a lower bound of asymptotic worst - case performance ratio , another problem is scheduling independent parallel tasks in parallel identical machine systems to minimize the makespan , we use strip packing method for it and give an approximation algorithm with asymptotic performance ratio no more than 1 . 6
第一个算法利用装箱问题中的ff ( firstfit )算法求解极大化按期完工工件数的平行机排序问题pm d _ j = d n - u _ j ，该近似算法具有渐近性能比下界。第二个近似算法利用二维装箱中strippacking问题的算法求解以极小化makespan为目标的带并行工件的平行机排序问题，该算法的渐近性能比具有下界1 . 5和上界1 . 6 。 |
| 9. | It preestablish a parallel design pattern for each parallel project based on master / slave architecture , and the communication between computers is transparent to programmers . in another words , when programmers are developing parallel programs , they do n ' t need to learn any complicated primitives and write any codes related with communication between parallel tasks . just only set some parameters in the visual interfaces that provided by platform , and the codes will be generated by our system automatically and inserted to the fittest place in the files
系统为每一个并行计算工程设定了一个基于master slave架构的并行编程模型，并且使得各结点机之间的通信对于开发人员透明化，也就是说，开发人员在开发并行程序时，无需了解纷繁复杂的通信原语，不涉及任务之间通信的代码，只需要在并行程序开发平台提供的可视化界面上对数据的传送进行设定，通信代码将自动生成并插入到程序合适的地方。 |
