Memory allocation method and delay perception-memory allocation apparatus suitable for memory access delay balance among multiple nodes in NUMA construction

Abstract

The present invention discloses a memory allocation method and delay perception-memory allocation apparatus suitable for memory access delay balance among multiple nodes in an NUMA construction. The apparatus comprises a delay perception unit (1) embedded inside a GQ unit of the NUMA multi-core construction, and a memory allocation unit (2) embedded inside a Linux operating system. According to the memory allocation method disclosed by the present invention, memory access delay among nodes in memory can be perceived periodicallyby the delay perception unit (1), whether memory access delay among nodes in memory is balanced can be obtained by the memory allocation unit (2), and a memory allocation node is selected according to a balance state, and is finally output to the Buddy memory allocator of the Linux operating system, thereby realizing physical memory allocation. According to the apparatus disclosed by the present invention aiming at an NUMA multi-core construction server, in the premise of ensuring memory access delay balance, application performance is stabilized, and unfairness of shared memory among application processes is reduced.

Description

technical field [0001] The present invention relates to a memory allocation method for a NUMA architecture server, more particularly, a memory allocation method suitable for memory access delay balance of a NUMA architecture server of a Linux operating system. Background technique [0002] With the popularization and development of multi-core architecture, NUMA (NonUniform Memory Access Architecture, non-uniform memory access architecture) multi-core architecture is widely adopted by major data centers and scientific computing clusters due to its advantages of low latency of local memory access. However, the complex structure of NUMA makes memory management more complicated for the operating system. Although the Linux operating system can make full use of the low-latency local memory access characteristics of the NUMA multi-core architecture, the problem of memory access delay balance among multiple memory nodes has not been resolved. How to effectively manage the use of me...

AI Technical Summary

Problems solved by technology

The NUMA multi-core architecture alleviates the contention of multiple cores for the same memory controller (IMC), but the unbalanced memory access delay between multiple memory nodes leads to the imbalance of shared memory resources between multiple running application processes (process). Fairness, which in turn increases the difference in process application performance (process performance), resulting in fluctuations in overall application performance (application performance)

The memory allocation method of the Linux operating system only considers the memory capacity (memory capacity) that can be allocated by memory nodes when allocating memory, which will cause the imbalance of memory access latency (memory access latency) between memory nodes

Images