摘要:
A data layout optimization may utilize affinity estimation between paris of fields of a record in a computer program. The affinity estimation may be determined based on a trace of an execution and in view of actual processing entities performing each access to the fields. The disclosed subject matter may be configured to be aware of a specific architecture of a target computer having a plurality of processing entities, executing the program so as to provide an improved affinity estimation which may take into account both false sharing issues, spatial locality improvement and the like.
摘要:
Dynamic determination of affinity between fields of structure may be determined based on accesses to the same instance. The affinity may be utilized in determining a data layout of a structure so as to optimize performance of a target program. The affinity determination may be an estimation based upon a trace of an execution of the target program. Access relation between proximate accesses to fields of the same instance may be utilized to estimate an optimized data layout of the structure.
摘要:
A data layout optimization may utilize affinity estimation between pairs of fields of a record in a computer program. The affinity estimation may be determined based on a trace of an execution and in view of actual processing entities performing each access to the fields. The disclosed subject matter may be configured to be aware of a specific architecture of a target computer having a plurality of processing entities, executing the program so as to provide an improved affinity estimation which may take into account both false sharing issues, spatial locality improvement and the like.
摘要:
A method and apparatus for performing efficient interprocess communication (IPC) in a computer system. With this invention, a memory region called the IPC transfer region is shared among all processes of the system to enable more efficient IPC. The unique physical address of the region is mapped into a virtual address from each of the address spaces of the processes of the system. When one of the processes needs to transfer data to another of the processes, the first process stores arguments describing the data in the region using the virtual address in its address space that maps into the unique physical address. When the other or second process needs to receive the data, the second process reads the data from the second region using the virtual address in its memory space that maps into the unique physical address. With this invention, in most cases, control of the IPC transfer region occurs automatically without any kernel intervention.
摘要:
In a computer system, a method and apparatus for scheduling activities' access to a resource with minimal involvement of the kernel of the operating system. More specifically, a “next bid” is maintained, and this parameter identifies the highest bid for the resource by any activity not currently accessing the resource. The accessing activity then compares its bid, which can be time varying, with the “next bid” to determine whether it should release the resource to another activity. The “next bid” can be accessed without any system calls to the operating system. This allows the activity to determine whether to relinquish control to the system without the necessity of communication between the two. Likewise, the operating system can access the bid of the accessing activity without explicit communication. This allows the system to determine whether to preempt the accessing activity without the necessity of communication between the two.