Data access accelerator
Various systems, methods, and processes for accelerating data access in application and testing environments are disclosed. A production dataset is received from a storage system, and cached in a consolidated cache. The consolidated cache is implemented by an accelerator virtual machine. A file system client intercepts a request for the production dataset from one or more application virtual machines, and transmits the request to the accelerator virtual machine. The accelerator virtual machine serves the production dataset to the one or more application virtual machines from the consolidated cache.
1. A computer-implemented method comprising:
obtaining a copy of a production dataset from a shared storage system, wherein the copy is received by an accelerator virtual machine,
the accelerator virtual machine is hosted by a virtualization server,
a plurality of application virtual machines are communicatively coupled to the accelerator virtual machine,
the copy comprises data that is common to the plurality of application virtual machines (“common data”) that are communicatively coupled to the accelerator virtual machine, and
the copy is stored in a consolidated cache of the accelerator virtual machine;
intercepting, by a file system client, a request for requested data, wherein
the request is received from a first application virtual machine of the plurality of application virtual machines,
the request is directed from the first application virtual machine to a storage system, and
the requested data comprises
a subset of the common data, and
data that has been changed by the first application virtual machine (“changed data”) is recorded in a writeable overlay, wherein
the writeable overlay comprises information that is specific to the first application virtual machine but is not associated with any other application virtual machine of the plurality of application virtual machines;
forwarding the request from the file system client to the accelerator virtual machine, wherein
the file system client does not forward the request to the storage system; and
serving, to the first application virtual machine, a first portion of the requested data from the consolidated cache of the accelerator virtual machine, and a second portion of the requested data from the writeable overlay, wherein
the first portion of the requested data comprises the subset of the common data, and
the second portion of the requested data comprises the changed data.
2. The computer-implemented method of claim 1 , wherein
receiving the production dataset and caching the common data are managed by the accelerator virtual machine.
3. The computer-implemented method of claim 1 , wherein
the production dataset comprises common production data required by the plurality of application virtual machines, and
the production dataset is a backup copy or a cover snapshot of a live production dataset.
4. The computer-implemented method of claim 1 , wherein the plurality of application virtual machines, the accelerator virtual machine, the consolidated cache, and the file system client are supported by a hypervisor.
5. The computer-implemented method of claim 1 , wherein
each of the plurality of application virtual machines employs one or more thin-provisioned virtual disks, and
each of the one or more thin-provisioned virtual disks store changes to the production dataset made by the corresponding application virtual machine.
6. The computer-implemented method of claim 1 , wherein
the storage system is part of a production environment, and
the plurality of application virtual machines, the accelerator virtual machine, the consolidated cache, and the file system client are part of an application development and testing environment.
7. The method of claim 1 , further comprising:
configuring a writeable overlay for each of the plurality of application virtual machines.
8. The method of claim 7 , wherein
the requested data comprises common data requested by the plurality of application virtual machines, and
the writeable overlay maintains changed data requested by the plurality of virtual machines.
9. The method claim 8 , further comprising:
intercepting another request for requested data from one or more application virtual machines; and
serving to the one or more application virtual machines a first portion of the requested data comprising the common data from the consolidated cache and a second portion of the requested data comprising the changed data from the corresponding writeable overlay configured for each of the one or more application virtual machines.
10. The method of claim 1 , further comprising:
requesting a copy of the production dataset from the storage system; and
caching the copy of the production dataset in the consolidated cache.
11. The method of claim 1 , wherein
the data in the consolidated cache is not split among the one or more application virtual machines.
12. A system comprising:
one or more processors; and
a memory coupled to the one or more processors, wherein the memory stores program instructions executable by the one or more processors to:
obtain a copy of a production dataset from a shared storage system, wherein the copy is received by an accelerator virtual machine,
the accelerator virtual machine is hosted by a virtualization server,
a plurality of application virtual machines are communicatively coupled to the accelerator virtual machine,
the copy comprises data that is common to the plurality of application virtual machines (“common data”) that are communicatively coupled to the accelerator virtual machine, and
the copy is stored in a consolidated cache of the accelerator virtual machine;
intercept, by a file system client, a request for requested data, wherein
the request is received from a first application virtual machine of the plurality of application virtual machines,
the request is directed from the first application virtual machine to a storage system, and
the requested data comprises
a subset of the common data, and
data that has been changed by the first application virtual machine (“changed data”) is recorded in a writeable overlay, wherein
the writeable overlay comprises information that is specific to the first application virtual machine but is not associated with any other application virtual machine of the plurality of application virtual machines;
forward the request from the file system client to the accelerator virtual machine, wherein
the file system client does not forward the request to the storage system; and
serve, to the first application virtual machine, a first portion of the requested data from the consolidated cache of the accelerator virtual machine, and a second portion of the requested data from the writeable overlay, wherein
the first portion of the requested data comprises the subset of the common data, and
the second portion of the requested data comprises the changed data.
13. The system of claim 12 , wherein
receiving the production dataset and caching the common data are managed by the accelerator virtual machine.
14. The system of claim 12 , wherein
the production dataset comprises common production data required by the plurality of application virtual machines, and
the production dataset is a backup copy or a cover snapshot of a live production dataset.
15. The system of claim 12 , wherein the plurality of application virtual machines, the accelerator virtual machine, the consolidated cache, and the file system client are supported by a hypervisor.
16. The system of claim 12 , wherein
each of the plurality of application virtual machines employs one or more thin-provisioned virtual disks, and
each of the one or more thin-provisioned virtual disks store changes to the production dataset made by the corresponding application virtual machine.
17. A non-transitory computer readable storage medium comprising program instructions executable to:
obtain a copy of a production dataset from a shared storage system, wherein the copy is received by an accelerator virtual machine,
the accelerator virtual machine is hosted by a virtualization server,
a plurality of application virtual machines are communicatively coupled to the accelerator virtual machine,
the copy comprises data that is common to the plurality of application virtual machines (“common data”) that are communicatively coupled to the accelerator virtual machine, and
the copy is stored in a consolidated cache of the accelerator virtual machine;
intercept, by a file system client, a request for requested data, wherein
the request is received from a first application virtual machine of the plurality of application virtual machines,
the request is directed from the first application virtual machine to a storage system, and
the request comprises
a subset of the common data, and
data that has been changed by the first application virtual machine (“changed data”) is recorded in a writeable overlay, wherein
the writeable overlay comprises information that is specific to the first application virtual machine but is not associated with any other application virtual machine of the plurality of application virtual machines;
forward the request from the file system client to the accelerator virtual machine, wherein
the file system client does not forward the request to the storage system; and
serve, to the first application virtual machine, a first portion of the requested data from the consolidated cache of the accelerator virtual machine, and a second portion of the requested data from the writeable overlay, wherein
the first portion of the requested data comprises the subset of the common data, and
the second portion of the requested data comprises the changed data.
18. The non-transitory computer readable storage medium of claim 17 , wherein
receiving the production dataset and caching the common data are managed by the accelerator virtual machine.
19. The non-transitory computer readable storage medium of claim 17 , wherein
the production dataset comprises common production data required by the plurality of application virtual machines, and
the production dataset is a backup copy or a cover snapshot of a live production dataset.
20. The non-transitory computer readable storage medium of claim 17 , wherein the plurality of application virtual machines, the accelerator virtual machine, the consolidated cache, and the file system client are supported by a hypervisor.
21. The non-transitory computer readable storage medium of claim 17 , wherein
each of the plurality of application virtual machines employs one or more thin-provisioned virtual disks, and
each of the one or more thin-provisioned virtual disks store changes to the production dataset made by the corresponding application virtual machine.
22. The non-transitory computer readable storage medium of claim 17 , wherein
the storage system is part of a production environment, and
the plurality of application virtual machines, the accelerator virtual machine, the consolidated cache, and the file system client are part of an application development and testing environment.