Accelerated data operations

Abstract

Systems, methods, and software described herein facilitate accelerated input and output operations with respect to virtualized environments. In an implementation, upon being notified of a guest read process initiated by a guest element running in a virtual machine to read data into a location in guest memory associated with the guest element, a computing system identifies a location in host memory associated with the location in the guest memory and initiates a host read process to read the data into the location in the host memory that corresponds to the location in the guest memory.

What is claimed is: 
     
         1 . An apparatus comprising:
 one or more non-transitory computer readable storage media; and   program instructions stored on the one or more non-transitory computer readable media, wherein the program instructions comprise a kernel driver that, when executed by a processing system, directs the processing system to at least:
 obtain a process identifier for a process running on a host; 
 identify host memory allocated to the process using the process identifier; 
 identify a portion of the host memory allocated to a virtual machine associated with the process; 
 map a portion of the host memory allocated to the virtual machine to a translation node running on the host. 
   
     
     
         2 . The apparatus of  claim 1  wherein the program instructions further comprise the translations node, wherein the translation node, when executed by the processing system, directs the processing system to at least:
 upon being notified of a guest read process initiated by a guest element running in the virtual machine to read data into a location in guest memory associated with the guest element, identify a location in the portion of the host memory mapped to the location in the guest memory; and 
 initiate a host read process to read the data into the location in the portion of the host memory mapped to the location in the guest memory. 
 
     
     
         3 . The apparatus of  claim 2  wherein the process comprises a hypervisor process. 
     
     
         4 . The apparatus of  claim 3  wherein the hypervisor process comprises a QEMU process. 
     
     
         5 . The apparatus of  claim 4  wherein the virtual machine is launched by the QEMU process. 
     
     
         6 . The apparatus of  claim 2  wherein the guest element comprises an operating system of the virtual machine. 
     
     
         7 . The apparatus of  claim 2  wherein the guest element comprises a process executing within an operating system of the virtual machine. 
     
     
         8 . An apparatus comprising:
 one or more storage devices;   a processing system operatively coupled with the one or more storage devices; and   program instructions stored on the one or more storage devices, wherein the program instructions comprise a kernel driver that, when executed by a processing system directs the processing system to at least:
 obtain a process identifier for a process running on a host; 
 identify host memory allocated to the process using the process identifier; 
 identify a portion of the host memory allocated to a virtual machine associated with the process; 
 map a portion of the host memory allocated to the virtual machine to a translation node running on the host. 
   
     
     
         9 . The apparatus of  claim 8  wherein the program instructions further comprise the translation node, wherein the translation node, when executed by the processing system, directs the processing system to at least:
 upon being notified of a guest read process initiated by a guest element running in the virtual machine to read data into a location in guest memory associated with the guest element, identify a location in the portion of the host memory mapped to the location in the guest memory; and 
 initiate a host read process to read the data into the location in the portion of the host memory mapped to the location in the guest memory. 
 
     
     
         10 . The apparatus of  claim 9  wherein the process comprises a hypervisor process; 
     
     
         11 . The apparatus of  claim 10  wherein the hypervisor process comprises a QEMU process. 
     
     
         12 . The apparatus of  claim 11  wherein the virtual machine is launched by the QEMU process. 
     
     
         13 . The apparatus of  claim 9  wherein the guest element comprises an operating system of the virtual machine. 
     
     
         14 . The apparatus of  claim 9  wherein the guest element comprises a process executing within an operating system of the virtual machine. 
     
     
         15 . A method comprising:
 in a kernel driver:
 obtaining a process identifier for a process running on a host; 
 identifying host memory allocated to the process using the process identifier; 
 identifying a portion of the host memory allocated to a virtual machine associated with the process; 
 mapping a portion of the host memory allocated to the virtual machine to a different process running on the host. 
   
     
     
         16 . The method of  claim 15  further comprising the different process:
 upon being notified of a guest read process initiated by a guest element running in the virtual machine to read data into a location in guest memory associated with the guest element, identifying a location in the portion of the host memory mapped to the location in the guest memory; and 
 initiating a host read process to read the data into the location in the portion of the host memory mapped to the location in the guest memory. 
 
     
     
         17 . The method of  claim 16  wherein the process comprises a hypervisor process; 
     
     
         18 . The method of  claim 17  wherein the hypervisor process comprises a QEMU process. 
     
     
         19 . The method of  claim 16  wherein the guest element comprises an operating system of the virtual machine. 
     
     
         20 . The method of  claim 16  wherein the guest element comprises a process executing within an operating system of the virtual machine.