High-Speed Data Storage

• US 20150286420A1
• Filed: 04/03/2014
• Published: 10/08/2015
• Est. Priority Date: 04/03/2014
• Status: Active Grant

• Current Assignee: Analysis Solutions, LLC
• Original Assignee: Analysis Solution LLC
• Inventors: Crisman, Jayson B., Tompkins, Ray D.
• Granted Patent: US 9,619,157 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0613 in relation to throughput
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0655 Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
• CPC Class Codes: G06F 3/0674 Disk device
• CPC Class Codes: G06F 3/0677 Optical disk device, e.g. CD-ROM, DVD
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD

Abstract

System and method for high speed data storage. An unmounted unformatted direct access storage device (DASD) may be opened via an operation system (OS) application programming interface (API). A maximum amount of storage space afforded by the DASD may be determined, and the DASD locked, thereby limiting write access to the DASD to a first application process and excluding writes to the DASD by any other application process. A single data type data stream is received, and stored on the DASD in linear fashion as it is received without using a file system architecture, metadata, indices, or keys associated with the data, including writing a block of the stream of data to the DASD and returning a next contiguous write location on the DASD in an iterative manner. The stored data are readable by multiple readers in a linear manner without using the file system architecture, metadata, indices, or keys.

Claims

1. A method for storing data, comprising:

a computer performing;

opening an unmounted unformatted direct access storage device (DASD) via an operation system (OS) application programming interface (API);

determining a maximum amount of storage space afforded by the DASD;

locking the DASD, thereby limiting write access to the DASD to a first application process and excluding writes to the DASD by any other application process;

receiving a stream of data, wherein the data are of a single data type, and wherein said receiving is invoked by the first application process; and

storing the stream of data on the DASD in a linear manner as it is received, wherein said storing does not use a file system architecture, metadata, indices, or keys associated with the data, wherein said storing comprises;

writing a block of the stream of data to the DASD;

returning a next contiguous write location on the DASD; and

performing said writing and said returning multiple times in an iterative manner;

wherein the stored data are readable by multiple readers in a linear manner without using the file system architecture, metadata, indices, or keys.

2. The method of claim 1, further comprising:

placing a marker at the beginning of the DASD, thereby identifying the DASD and establishing an exclusive relationship between the first application process and the DASD.

3. The method of claim 1, wherein said performing said writing and said returning multiple times in an iterative manner comprises:

wrapping back to the beginning of the DASD in response to the DASD being full.

4. The method of claim 3, wherein said wrapping comprises:

indicating, by the next contiguous write location, the beginning of the DASD in response to the DASD being full; and

writing an immediately subsequent block of the stream of data at the beginning of the DASD.

5. The method of claim 1, further comprising:

halting said performing said writing and said returning in response to the DASD being full.

6. The method of claim 1, wherein the stream of data comprises network packets.

7. The method of claim 1, wherein the stream of data comprises geological data.

8. The method of claim 1, wherein said storing comprises:

converting the received stream of data into blocks of data; and

storing each block of data with a respective sequential write to the DASD;

wherein said determining the maximum amount of storage space afforded by the DASD comprises determining a maximum number of blocks of data afforded by the DASD, and wherein each next contiguous write location is a raw block address on the DASD.

9. The method of claim 8, wherein said storing the stream of data on the DASD in a linear manner is performed by a software driver configured to optimize speed and throughput of the DASD for storing sequential data.

10. The method of claim 9, wherein to optimize speed and throughput of the DASD for storing sequential data, the software driver stores the blocks of data on the DASD at contiguous locations on the DASD, thereby minimizing mechanical movement of the DASD.

11. The method of claim 1, wherein the DASD is configured for access by one or more linear read processes, and wherein each linear read process is restricted to reading data from the DASD starting at the beginning of the DASD and reading the data in a linear manner.

12. A system for storing data, comprising:

a computer, comprising;

a processor; and

a memory, coupled to the processor, wherein the memory stores program instructions; and

a direct access storage device (DASD), coupled to the computer, wherein the DASD is initially unmounted and unformatted;

wherein the program instructions are executable by the processor to;

open the unmounted unformatted DASD via an operation system (OS) application programming interface (API);

determine a maximum amount of storage space afforded by the DASD;

lock the DASD, thereby limiting write access to the DASD to a first application process and excluding writes to the DASD by any other application process;

receive a stream of data, wherein the data are of a single data type, and wherein said receiving the stream of data is invoked by the first application process; and

store the stream of data on the DASD in a linear manner as it is received, wherein said storing does not use a file system architecture, metadata, indices, or keys associated with the data, wherein to store the stream of data on the DASD, the program instructions are executable to;

write a block of the stream of data to the DASD;

return a next contiguous write location on the DASD; and

perform said writing and said returning multiple times in an iterative manner;

wherein the stored data are readable by multiple readers in a linear manner without using the file system architecture, metadata, indices, or keys.

13. The system of claim 12, wherein the program instructions are further executable to:

place a marker at the beginning of the DASD, thereby identifying the DASD and establishing an exclusive relationship between the first application process and the DASD.

14. The system of claim 12, wherein to implement said performing said writing and said returning multiple times in an iterative manner, the program instructions are executable to:

wrap back to the beginning of the DASD in response to the DASD being full.

15. The system of claim 14, wherein to wrap back to the beginning of the DASD in response to the DASD being full, the program instructions are executable to:

indicate, by the next contiguous write location, the beginning of the DASD in response to the DASD being full; and

write an immediately subsequent block of the stream of data at the beginning of the DASD.

16. The system of claim 12, wherein the program instructions are further executable to:

halt said performing said writing and said returning in response to the DASD being full.

17. The system of claim 12, wherein the stream of data comprises network packets.

18. The system of claim 12, wherein the stream of data comprises geological data.

19. The system of claim 12, wherein to store the stream of data on the DASD in a linear manner, the program instructions are executable to:

convert the received stream of data into blocks of data; and

store each block of data with a respective sequential write to the DASD;

wherein said determining the maximum amount of storage space afforded by the DASD comprises determining a maximum number of blocks of data afforded by the DASD, and wherein each next contiguous write location is a raw block address on the DASD.

20. The system of claim 19, wherein the program instructions implement a software driver configured to optimize speed and throughput of the DASD for storing sequential data, and wherein the software driver is executable to perform said storing the stream of data on the DASD in a linear manner.

21. The system of claim 20, wherein to optimize speed and throughput of the DASD for storing sequential data, the software driver stores the blocks of data on the DASD at contiguous locations on the DASD, thereby minimizing mechanical movement of the DASD.

22. The system of claim 12, wherein the DASD is configured for access by one or more linear read processes, and wherein each linear read process is restricted to reading data from the DASD starting at the beginning of the DASD and reading the data in a linear manner.

23. A non-transitory computer accessible memory medium that stores program instructions executable by a processor to:

open an unmounted unformatted direct access storage device (DASD) via an operation system (OS) application programming interface (API);

determine a maximum amount of storage space afforded by the DASD;

lock the DASD, thereby limiting write access to the DASD to a first application process and excluding writes to the DASD by any other application process;

receive a stream of data, wherein the data are of a single data type, and wherein said receiving the stream of data is invoked by the first application process; and

store the stream of data on the DASD in a linear manner as it is received, wherein said storing does not use a file system architecture, metadata, indices, or keys associated with the data, wherein to store the stream of data on the DASD, the program instructions are executable to;

write a block of the stream of data to the DASD;

return a next contiguous write location on the DASD; and

perform said writing and said returning multiple times in an iterative manner;

wherein the stored data are readable by multiple readers in a linear manner without using the file system architecture, metadata, indices, or keys.