Part of the solution is to implement per-CPU scheduling queues.
这个问题的解决方案是实现单cpu调度队列。
The kernel provides other functions for per-CPU locking and dynamic allocation of variables.
内核提供了用于per - CPU锁定和变量动态分配的其他函数。
If per-CPU data is needed (as determined in the section header checks), a per-CPU block is allocated.
如果需要per - CPU数据(这在检查区段头时确定)，那么就分配per - cpu块。
Locking should ensure both that per-CPU data structures and state are always protected against preemption.
锁定可以确保每个CPU的数据结构和状态始终受到保护而不被抢占。
Defining a per-CPU variable is done with the DEFINE_PER_CPU macro, to which you provide a type and variable name.
per - CPU变量由define_per_cpu宏进行定义，需要为该宏提供类型和变量名称。
For this reason, the 2.6 kernel introduced the concept of per-CPU variables that are associated with a single CPU.
出于此原因，2.6内核引入了per - CPU变量的概念，这些变量与单个CPU相关联。
The mpstat command reports CPU utilization on a per-CPU basis, and it reports extended statistics, including context switches, interrupts, and so on.
mpstat命令报告每个CPU的使用情况，也报告延伸的统计信息，包括上下文切换、中断等等。
The new scheduler addresses these issues by distributing timeslices on a per-CPU basis and eliminating the global synchronization and recalculation loop.
新的调度器解决上述问题的方法是，基于每个 CPU 来分布时间片，并且取消了全局同步和重算循环。
Tasklets are scheduled through the softirq mechanism, sometimes through ksoftirqd (a per-CPU kernel thread), when the machine is under heavy soft-interrupt load.
通过软中断机制来调度微线程，当机器处于严重软件中断负荷之下时， 可通过ksoftirqd（一种每CPU内核线程）软中断来调度。
The macro creates an array of variables, one per CPU instance.
该宏创建了一个变量数组（每个 CPU一个变量）。
You can use the environment variable KAIOON to control the number of requests allocated per CPU virtual processor.
您可以使用环境变量KAIOON来控制分配给每个CPU虚拟处理器的请求数量。
The table below shows the CPU cost (in milliseconds per message) for different accounting and statistics collection settings.
下表显示了不同的计算和统计收集设置的CPU成本（微秒每消息）。
Threading and the number of processes per CPU Settings are configurable by administrators.
每个CPU上多线程和进程数的设置可由管理配置。
The CPU speed (in milliseconds per instruction) is used by the SQL optimizer to estimate the cost of performing certain operations.
SQL优化器使用CPU速度（每条指令几微秒）来评估某些操作的执行成本。
The measurement job puts 1000 messages, so the CPU cost is 52.3 milliseconds per message.
测量任务放置了1000条消息，所以每条消息的CPU成本是52.3微秒。
The cost can be one of the following metrics: execution time per query execution, user CPU time per query execution, and system CPU time per query execution.
这里所说的成本可能是下面的度量之一：每次查询执行的执行时间、每次查询执行的用户CPU时间、以及每次查询执行的系统CPU时间。
If you have multiple CPUs, it will provide this information per CPU and the system-wide average of all CPUs.
如果您有多个CPU，它将给出每个CPU的统计数据以及系统中所有CPU的平均统计数据。
For instance, the company says developers should expect to pay 10 to 12 cents per CPU core-hour.
举例来说，该公司表示开发者每CPU核心/小时可能会需要支付10到12美分。
The next column shows total CPU cost (in milliseconds per message).
接下来的一列显示总的cpu成本(微秒每消息)。
Figure 5 shows the CPU cost per user for MLC83 and MLC96 over the past year.
图5显示了过去一年针对 MLC83和MLC96的每用户成本。
It is very close to the CPU cost of 52.21 milliseconds per message.
这非常接近每条消息的CPU成本52.21微秒。
This provides fair access for all tasks to the CPU (and locking only on a per CPU basis).
这就为所有的任务提供了公平访问CPU的机会(仅根据每个CPU锁定)。
AIX automatically tries to encourage processor affinity by having one run queue per CPU, which was discussed earlier.
AIX会在每个CPU上都设置一个运行队列，以此来自动尝试推动处理器关联性的建立，这一点我在前面已经讨论过。
The total CPU cost per message, taking capture ratio into account is, in this case.
把捕获比考虑进去后，在这种情况下，每条消息的总CPU成本是。
For planning purposes, we suggest that an environment should be sized to accommodate 50 Concurrent users per modern CPU, regardless of the user role.
为达成计划的目的，我建议一个环境的大小应该控制在每个现代CPU50个Concurrent用户，不考虑用户角色。
Doubled the number of jobs per node, benefitting from increased CPU power due to Moore's law.
受惠于摩尔定律带来的CPU计算能力增加，每个节点的任务数翻了一倍。
These results demonstrate that Domino 7 supports a larger number of users in a single partition, and also lowers CPU requirements per user.
结果表明Domino7在单一分区中能支持更多的用户，同时还降低了每个用户的CPU需求。
This translates into eight concurrent interactive reporting requests per physical CPU.
这就是说每个物理cpu有8个并发交互式报告请求。
This translates into four concurrent batch reports executed simultaneously per physical CPU.
这就是说每个物理CPU同时可以执行4 个并发批处理报告请求。
This translates into four concurrent batch reports executed simultaneously per physical CPU.
这就是说每个物理CPU同时可以执行4 个并发批处理报告请求。
应用推荐