System and method of balancing mixed workload performance

1. A computer implemented method of balancing mixed workload performance, the method comprising:

monitoring compression and decompression workload at a hardware accelerator using a hardware accelerator quality of service (QoS) scheduler;

monitoring compression and decompression workload at a central processing unit (CPU) using a CPU QoS scheduler, wherein the compression and decompression workload includes one or more of backup jobs, restore jobs, garbage collection jobs, and replication jobs;

comparing the workload at the hardware accelerator and the workload at the CPU; and

allocating tasks between the hardware accelerator and the CPU to obtain an optimal bandwidth at the hardware accelerator and the CPU.

2. The computer implemented method as in claim 1 , wherein the hardware accelerator is a QuickAssist Technology gzip accelerator.

3. The computer implemented method as in claim 1 , further comprising:

receiving workload priorities from a user prioritizing the backup jobs, restore jobs, garbage collection jobs, and replication jobs.

4. The computer implemented method as in claim 1 , wherein the optimal bandwidth includes a balanced bandwidth between the hardware accelerator and the CPU.

5. The computer implemented method as in claim 1 , further comprising:

aggregating multiple source data buffers into a compression unit;

submitting a scatter gather list and a destination buffer to the hardware accelerator;

calculating a first checksum of result data from the destination buffer;

adding a zlib header to the result data; and

adding the first checksum as a zlib footer to the result data.

6. A data compression system comprising:

a hardware accelerator for compressing and decompressing data;

a hardware accelerator quality of service (QoS) scheduler configured to monitor workload of and assign compression and decompression jobs to the hardware accelerator;

a central processing unit (CPU) QoS scheduler configured to monitor workload of a CPU and assign compression and decompression jobs to the CPU; and

mixed workload performance logic configured to:

monitor compression and decompression workload at the hardware accelerator using the hardware accelerator QoS scheduler;

monitor compression and decompression workload at the CPU using the CPU QoS scheduler;

compare the workload at the hardware accelerator and the workload at the CPU, wherein the compression and decompression workload at the hardware accelerator and at the CPU includes one or more of backup jobs, restore jobs, garbage collection jobs, and replication jobs; and

allocate tasks between the hardware accelerator and the CPU to obtain an optimal bandwidth at the hardware accelerator and the CPU.

7. The data compression system as in claim 6 , wherein the hardware accelerator is a QuickAssist Technology gzip accelerator.

8. The data compression system as in claim 6 , wherein the mixed workload performance logic is further configured to:

receive workload priorities from a user prioritizing the backup jobs, restore jobs, garbage collection jobs, and replication jobs.

9. The data compression system as in claim 6 , wherein the optimal bandwidth includes a balanced bandwidth between the hardware accelerator and the CPU.

10. The data compression system as in claim 6 , further comprising compression logic configured to:

aggregate multiple source data buffers into a compression unit;

submit a scatter gather list and a destination buffer to the hardware accelerator;

calculate a first checksum of result data from the destination buffer;

add a zlib header to the result data; and

add the first checksum as a zlib footer to the result data.

11. A non-transitory computer-readable medium storing instructions which, when executed by one or more processors, cause the one or more processors to perform a mixed workload performance operation, the operation comprising:

monitoring compression and decompression workload at a hardware accelerator using a hardware accelerator QoS scheduler;

monitoring compression and decompression workload at a central processing unit (CPU) using a CPU QoS scheduler;

comparing the workload at the hardware accelerator and the workload at the CPU, wherein the compression and decompression workload at the hardware accelerator and at the CPU includes one or more of backup jobs, restore jobs, garbage collection jobs, and replication jobs; and

allocating tasks between the hardware accelerator and the CPU to obtain an optimal bandwidth at the hardware accelerator and the CPU.

12. The non-transitory computer-readable medium as in claim 11 , wherein the hardware accelerator is a QuickAssist Technology gzip accelerator.

13. The non-transitory computer-readable medium as in claim 11 , wherein the mixed workload performance operation is further configured to:

receive workload priorities from a user prioritizing the backup jobs, restore jobs, garbage collection jobs, and replication jobs.

14. The non-transitory computer-readable medium as in claim 11 , wherein the optimal bandwidth includes a balanced bandwidth between the hardware accelerator and the CPU.

15. The non-transitory computer-readable medium as in claim 11 , wherein the one or more processors are further configured to perform a compression operation to:

aggregate multiple source data buffers into a compression unit;

submit a scatter gather list and a destination buffer to the hardware accelerator;

calculate a first checksum of result data from the destination buffer;

add a zlib header to the result data; and

add the first checksum as a zlib footer to the result data.