Deploying software in a multi-instance node

摘要

A system for deploying big data software in a multi-instance node. The optimal CPU memory and core configuration for a single instance database is determined. After determining an optimal core-memory ratio for a single instance execution, the software is deployed in multi-instance mode on single machine by applying the optimal core-memory ratio for each of the instances. The multi-instance database may then be deployed and data may be loaded in parallel for the instances.

主权项

1. A method for identifying a preferred central processing unit (CPU) core to memory ratio, the method comprising:
executing instructions that correspond to a workload of an application program out of a memory by a first CPU core from a plurality of CPU cores; identifying an amount of memory used by the first CPU core when the first CPU core executes the corresponding instructions of the application program out of the memory; executing the corresponding instructions out of the memory by two or more CPU cores of the plurality of CPU cores out of the memory; identifying an initial measure of CPU usage when the two or more cores of the plurality of CPU cores execute the corresponding instructions; changing the number of CPU cores of the plurality of CPU cores that Mon execute the corresponding instructions; executing the corresponding instructions by the changed number of CPU cores; identifying a subsequent measure of CPU usage when the changed number of CPU cores execute the corresponding instructions; identifying a preferred number of CPU cores for executing the corresponding instructions based on comparing the initial measure of CPU usage and the subsequent measure of CPU usage to a CPU usage threshold level, wherein the CPU usage threshold level corresponds to a CPU core to memory ratio for executing the application program; and identifying the preferred CPU core to memory ratio, wherein the preferred CPU core to memory ratio also corresponds to the preferred number of CPU cores and to the identified amount of memory, wherein the preferred number of CPU cores and the identified amount of memory are allocated for executing the application program when at least one instance of the application program is subsequently deployed.