This section describes how VMware Cloud on AWS handles large memory pages.

In addition to the usual 4KB memory pages, VMware Cloud on AWS also provides 2MB memory pages (commonly referred to as “large pages”). (Note that although some CPUs support 1GB memory pages, VMware Cloud on AWS supports only 2MB large pages.)

VMware Cloud on AWS assigns these 2MB machine memory pages to guest operating systems whenever possible; it does this even if the guest operating system doesn’t request them (though the full benefit of large pages comes only when the guest operating system and applications use them as well, as described in Guest Operating System Memory Considerations). The use of large pages can significantly reduce TLB misses, improving the performance of most workloads, especially those with large active memory working sets. In addition, large pages can slightly reduce the per-virtual-machine memory space overhead.

Use of large pages can also change page sharing behavior. While VMware Cloud on AWS ordinarily uses page sharing regardless of memory demands, it does not share large pages. Therefore with large pages, page sharing might not occur until memory overcommitment is high enough to require the large pages to be broken into small pages. For further information see VMware KB article 78544. In VMware Cloud on AWS, disabling large pages can be done only on a per VM basis.