Synchronous replication end to end latency reporting

1. A data storage system, comprising:

a communication interface configured to receive, from a data storage system, performance data associated with processing and synchronously replicating file system operation requests; and

a processor coupled to the communication interface and configured to:

aggregate performance data associated with an application-level workload, wherein the performance data includes a first set of data associated with a primary storage system associated with the application-level workload and a secondary storage system associated with the application-level workload; and

provide, via a user interface, a representation of the aggregated performance data for the application-level workload, wherein:

the representation of the aggregated performance data includes a representation of a latency associated with the primary storage system and a representation of a latency associated with the secondary storage system, each representation of latency including factors contributing to each latency; and

at least one of the latency associated with the primary storage system and the latency associated with the secondary storage system includes a measured barrier latency corresponding to an amount of time that it takes to decide whether a given operation is to be replicated synchronously;

wherein the application-level workload comprises a virtual machine;

wherein the processor is further configured to:

create a mapping between the virtual machine and a set of one or more files, by tracking, at a storage system level, data indicating which files are associated with which virtual machines;

use said mapping to aggregate performance data associated with the one or more files;

store the aggregated performance data associated with the one or more files in a manner that associates the aggregated performance data associated with the one or more files with the virtual machine; and

wherein the stored aggregated performance data associated with the one or more files is used to provide at least said first set of data on a per virtual machine basis.

2. The data storage system of claim 1 , wherein the performance data includes data associated with performing, at the primary storage system, file system operations associated with the application-level workload.

3. The data storage system of claim 2 , wherein said data associated with performing, at the primary storage system, file system operations associated with the application-level workload includes one or more of: contention latency, flash access latency, disk access latency, and network transmission latency.

4. The data storage system of claim 1 , wherein the performance data includes data associated with performing file system operations associated with the application-level workload at the secondary storage system.

5. The data storage system of claim 1 , wherein the performance data includes data reflecting a latency associated with synchronously replicating, to the secondary storage system, file system operations associated with the application-level workload.

6. The data storage system of claim 5 , wherein the processor is further configured to determine that file system operation requests associated with the application-level workload are to be replicated synchronously to the secondary storage system based at least in part on the latency associated with synchronously replicating, to the secondary storage system, file system operations associated with the application-level workload.

7. The data storage system of claim 5 , wherein said latency associated with synchronously replicating, to the secondary storage system, file system operations associated with the application-level workload includes a file system operation splitting latency reflecting a difference in time between local completion of a file system operation at the primary storage system and receipt from the secondary storage system of an acknowledgement of completion of the file system operation at the secondary storage system.

8. The data storage system of claim 6 , wherein the latency associated with synchronously replicating, to the secondary storage system, file system operations associated with the application-level workload includes the measured barrier latency.

9. The data storage system of claim 1 , wherein the processor is further configured to track and report at least one of the measured barrier latency or a file system operation splitting latency.

10. A method, comprising:

receiving, from a data storage system via a communication interface, performance data associated with processing and synchronously replicating file system operation requests from a primary storage system to a secondary storage system;

aggregating performance data associated with an application-level workload, wherein the performance data includes a first set of data associated with a primary storage system associated with the application-level workload and a secondary storage system associated with the application-level workload; and

providing, via a user interface, a representation of the aggregated performance data for the application-level workload, wherein:

the representation of the aggregated performance data includes a representation of a latency associated with the primary storage system and a representation of a latency associated with the secondary storage system, each representation of latency including factors contributing to each latency; and

at least one of the latency associated with the primary storage system and the latency associated with the secondary storage system includes a measured barrier latency corresponding to an amount of time that it takes to decide whether a given operation is to be replicated synchronously,

wherein the application-level workload comprises a virtual machine;

creating a mapping between the virtual machine and a set of one or more files, by tracking, at a storage system level, data indicating which files are associated with which virtual machines;

using said mapping to aggregate performance data associated with the one or more files;

storing the aggregated performance data associated with the one or more files, in a manner that associates the aggregated performance data associated with the one or more files with the virtual machine; and

wherein the stored aggregated performance data associated with the one or more files is used to provide at least said first set of data on a per virtual machine basis.

11. The method of claim 10 , wherein the performance data includes data associated with performing, at the primary storage system, file system operations associated with the application-level workload.

12. The method of claim 11 , wherein the data associated with performing, at the primary storage system, file system operations associated with the application-level workload includes one or more of contention latency, flash access latency, disk access latency, and network transmission latency.

13. The method of claim 10 , wherein the performance data includes data associated with performing file system operations associated with the application-level workload at the secondary storage system.

14. The method of claim 10 , wherein the performance data includes data reflecting a latency associated with synchronously replicating, to the secondary storage system, file system operations associated with the application-level workload.

15. The method of claim 14 , wherein said latency associated with synchronously replicating, to the secondary storage system, file system operations associated with the application-level workload includes the measured barrier latency, and the measured barrier latency is associated with determining whether file system operation requests associated with the application-level workload are to be replicated synchronously to the secondary storage system.

16. A computer program product embodied in a non-transitory computer readable medium and comprising computer instructions for:

receiving, from a data storage system via a communication interface, performance data associated with processing and synchronously replicating file system operation requests from a primary storage system to a secondary storage system;

aggregating performance data associated with an application-level workload, wherein the performance data includes a first set of data associated with a primary storage system associated with the application-level workload and a secondary storage system associated with the application-level workload; and

providing, via a user interface, a representation of the aggregated performance data for the application-level workload, wherein:

the representation of the aggregated performance data includes a representation of a latency associated with the primary storage system and a representation of a latency associated with the second storage system, each representation of latency including factors contributing to each latency; and

at least one of the latency associated with the primary storage system and the latency associated with the secondary storage system includes a measured barrier latency corresponding to an amount of time that it takes to decide whether a given operation is to be replicated synchronously;

wherein the application-level workload comprises a virtual machine;

creating a mapping between the virtual machine and a set of one or more files, by tracking, at a storage system level, data indicating which files are associated with which virtual machines;

using said mapping to aggregate performance data associated with the one or more files;

storing the aggregated performance data associated with the one or more files, in a manner that associates the aggregated performance data associated with the one or more files with the virtual machine; and

wherein the stored aggregated performance data associated with the one or more files is used to provide at least said first set of data on a per virtual machine basis.