Reducing decompression latency in a compression storage system

Patent metadata

Field	Value
Publication number	US-9405762-B2
Application number	US-201414564743-A
Country	US
Kind code	B2
Filing date	Dec 9, 2014
Priority date	Jun 14, 2012
Publication date	Aug 2, 2016
Grant date	Aug 2, 2016

Abstract

In a compression processing storage system, using a pool of compression cores, the compression cores are assigned to process either compression operations, decompression operations, or decompression and compression operations, which are scheduled for processing. A minimal number of the compression cores are allocated for processing the compression operations, thereby increasing compression latency. Upon reaching a throughput limit for the compression operations that causes the minimal number of the plurality of compression cores to reach a busy status, the minimal number of the plurality of compression cores for processing the compression operations is increased.

First claim

What is claimed is:

1. A method for reducing decompression latency in a compression processing storage system using a pool of a plurality of compression cores, comprising:
assigning to the plurality of compression cores for processing only one of compression operations, decompression operations, and decompression and compression operations that are scheduled for processing; and
allocating a minimal number of compression cores for processing compression operations, thereby increasing a compression latency, wherein upon reaching a throughput limit for the compression operations that causes the minimal number of compression cores to reach a busy status, the minimal number of compression cores for processing the compression operations is increased.

2. The method of claim 1 , further including setting a maximum number of compression cores for processing only decompression operations, thereby lowering the decompression latency.

3. The method of claim 1 , further including scheduling the one of compression operations, decompression operations, and decompression and compression operations, between the pool of the plurality of processing cores, according to a thread weight value (TWV) that is assigned to each one of the plurality of processing cores having a difference in processing power.

4. The method of claim 1 , further including, if processing compression operations, sustaining a bandwidth of the compression operations by measuring core idle time and a throughput limit.

5. The method of claim 1 , further including setting a time-out period for a predetermined threshold period for processing the one of compression operations, decompression operations, and decompression and compression operations.

6. The method of claim 1 , further including, if processing compression and decompression operations, assigning the decompression and compression operations to each of the plurality of processing cores for processing if each of the plurality of compression cores have attained the busy status.

7. The method of claim 1 , wherein the busy status represents those of the plurality of compression cores that have processed the one of the compression operations, decompression operations, and decompression and compression operations during a predetermined threshold period.

8. The method of claim 1 , further including assigning only decompression operations to the plurality of compression cores having an idle status, wherein the idle status represents those of the plurality of compression cores that have yet to process the one of compression operations, decompression operations, and decompression and compression operations during a predetermined threshold period.

9. The method of claim 7 , further including:
if processing compression operations, determining whether each one of the plurality of compression cores had one of the idle status and the busy status during a last-predetermined threshold period; and
determining the minimum number of the plurality of compression cores by locating the lastmost positioned one of the plurality of compression cores to process the compression operations.

10. The method of claim 1 , further including determining the maximum number of the plurality of compression cores by, from a firstmost positioned one of the plurality of compression cores, sequentially counting, each one of the plurality of compression cores until reaching a lastmost positioned one of the plurality of compression cores, to include in the maximum number of the plurality of compression cores to process decompression operations.

11. The method of claim 9 , further including:
if processing compression operations, if the lastmost positioned one of the plurality of compression cores is determined to be in the busy status, and those of the plurality of compression cores to the left of positioned before the lastmost positioned one of the plurality of compression cores are in the idle status:
permitting one of the plurality of compression cores that is positioned immediately before the lastmost positioned one of the plurality of compression to process the compression operations.

12. The method of claim 11 , further including:
if processing compression operations, if at least one or more of an nth number of lastmost positioned ones of the plurality of compression cores are determined to be in the busy status, and all other of the plurality of
compression cores to the left of the at least one or more of the nth number of lastmost positioned ones of the plurality of compression cores are in the idle status:
permitting an additional one of the at least one or more of the nth number of lastmost positioned ones of the plurality of compression cores to process the compression operations.

13. A system for reducing decompression latency in a compression processing storage system using a pool of a plurality of compression cores, comprising:
at least one scheduler scheduling compression operations, decompression operations, and decompression and compression operations between the pool of the plurality of processing cores,
at least one compression cache, remotely connected to each of the plurality of compression cores, assisting with the compression operations;
at least one processor device, controlling the at least one scheduler and the at least one compression cache, and operable in the compression processing storage system, wherein the at least one processor device:
assigns to the plurality of compression cores for processing only one of compression operations, decompression operations, and decompression and compression operations that are scheduled for processing, and
allocates a minimal number of the plurality of compression cores for processing compression operations, thereby increasing a compression latency, wherein upon reaching a throughput limit for the compression operations that causes the minimal number of compression cores to reach a busy status, the minimal number of compression cores for processing the compression operations is increased.

14. The system of claim 13 , wherein the at least one processor device sets a maximum number of compression cores for processing only decompression operations, thereby lowering the decompression latency.

15. The system of claim 13 , wherein the at least one processor device assigns a thread weight value (TWV) to each one of the plurality of processing cores having a difference in processing power, wherein the compression operations, decompression operations, and decompression and compression operations, between the pool of the plurality of processing cores, are proportionally scheduled by the scheduler according to the TWV.

16. The system of claim 13 , wherein the at least one processor device sustains a bandwidth of the compression operations by measuring core idle time and a throughput limit.

17. The system of claim 13 , wherein the at least one processor device sets a time-out period for a predetermined threshold period for processing the compression operations, decompression operations, and decompression and compression operations.

18. The system of claim 13 , wherein the at least one processor device assigns the decompression and compression operations to each of the plurality of processing cores for processing if each of the plurality of compression cores have attained the busy status.

19. The method of claim 13 , wherein the busy status represents those of the plurality of compression cores that have processed at least one of the compression operations, decompression operations, and decompression and compression ope

Assignees

• IBM

Inventors

• Amit Jonathan
• Lidor Amir
• Marenkov Sergey
• Raikhman Rostislav

Classifications

• H04L47/70

  Admission control; Resource allocation · CPC title

• G06F16/1744

  using compression, e.g. sparse files · CPC title

• H03M7/6023 Primary

  Parallelization · CPC title

• H03M7/30

  Compression (speech analysis-synthesis for redundancy reduction G10L19/00; for image communication H04N); Expansion; Suppression of unnecessary data, e.g. redundancy reduction · CPC title

• G06F17/30153 Primary

  Physics · mapped topic