Method and system for record access in a distributed system

• US 9,489,336 B2
• Filed: 07/15/2013
• Issued: 11/08/2016
• Est. Priority Date: 07/15/2013
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Fuller, Benjamin J., Hueston, Robert J., Lach, Jorge E.
• Primary Examiner(s): Yen, Syling
• Application Number:
  US13/942,240
• Publication Number:
  US 20150019672A1
• Time in Patent Office: 1,212 Days
• Field of Search: 707/706, 707/713, 707/722, 707/736, 707/758, 707/781, 707/782, 707/783, 707/999.009, 707/999.01, 709/225, 709/229
• US Class Current: 1/1
• CPC Class Codes: G06F 15/167 using a common memory, e.g. mailbox
• CPC Class Codes: G06F 21/6227 where protection concerns the structure of data, e.g. records, types, queries
• CPC Class Codes: H04L 9/3228 One-time or temporary data, i.e. information which is sent for every authentication or authorization, e.g. one-time-password, one-time-token or one-time-key

Abstract

A method for record access in a distributed system includes receiving a request for a record, wherein the request comprises a transmitted key and a record identifier, extracting a location identifier and a transmitted pseudorandom portion from the transmitted key, obtaining a stored pseudorandom portion from a location in a key memory specified by the location identifier, and providing access to the record identified by the record identifier when the transmitted pseudorandom portion matches the stored pseudorandom portion.

Claims

1. A method for record access in a distributed system, comprising:

receiving, at a hardware gatekeeper from a requester node, a request for a record of a resource of a home node, wherein the request comprises a transmitted key and a record identifier;

extracting a location identifier and a transmitted pseudorandom portion from the transmitted key, wherein the location identifier specifies a location, in a key memory operatively connected to an authentication module, of a stored pseudorandom portion, and wherein the authentication module comprises the hardware gatekeeper;

obtaining the stored pseudorandom portion from the location in the key memory specified by the location identifier; and

providing the requester node access to the record identified by the record identifier when the transmitted pseudorandom portion matches the stored pseudorandom portion, wherein the transmitted pseudorandom portion and the stored pseudorandom portion are matched by the hardware gatekeeper;

processing, by the requester node, the record to obtain a first modified record;

storing the first modified record in a write buffer of a plurality of write buffers of the requester node;

obtaining a first storage status of each of the plurality of write buffers;

initializing, in a first scoreboard, a first plurality of entries corresponding to the plurality of write buffers based on the first storage status; and

for each write of the first modified record from the write buffer of the plurality of write buffers to the home node, clearing a first entry of the first plurality of entries based on the first entry corresponding to the write buffer,

wherein a first checkpoint is complete when each entry of the first plurality of entries in the first scoreboard is cleared.

2. The method of claim 1, further comprising:

obtaining auxiliary data from the key memory; and

processing the auxiliary data.

3. The method of claim 2, wherein the auxiliary data comprises:

a description of the resource accessible based on the transmitted key, wherein the resource comprises the record.

4. The method of claim 1, wherein the first scoreboard corresponds to a first thread and wherein the method further comprises:

after initializing the first plurality of entries and after a subset of the first plurality of entries is cleared;

reusing, for a second thread, at least one of the plurality of write buffers for storing a second modified record without modifying the first plurality of entries based on the storing of the second modified record;

obtaining a second storage status of each of the plurality of write buffers;

initializing, in a second scoreboard, a second plurality of entries corresponding to the plurality of write buffers based on the second storage status, wherein the second scoreboard corresponds to the second thread;

for each write of the second modified record from the write buffer of the plurality of write buffers to the home node, clearing a second entry of the second plurality of entries,

wherein a second checkpoint is complete when each entry of the second plurality of entries is cleared.

5. The method of claim 4, wherein the first scoreboard corresponds to a first plurality of threads comprising the first thread, and wherein the second scoreboard corresponds to a second plurality of threads comprising the second thread.

6. A system for record access in a distributed system, comprising:

an authentication module comprising;

key memory configured to store a plurality of pseudorandom portions of a plurality of keys; and

a hardware gatekeeper configured to;

receive a request for a record, wherein the request comprises a transmitted key and a record identifier;

extract a first location identifier and a transmitted pseudorandom portion from the transmitted key, wherein the first location identifier specifies a first location, in the key memory of a stored pseudorandom portion of the plurality of pseudorandom portions;

obtain the stored pseudorandom portion from the first location in the key memory specified by the first location identifier; and

provide access to the record identified by the record identifier by determining that the transmitted pseudorandom portion matches the stored pseudorandom portion;

a storage system for storing a plurality of write buffers;

a first scoreboard comprising a first plurality of entries, wherein each entry of the first plurality of entries corresponds to a write buffer of the plurality of write buffers; and

a first thread configured to;

obtain a first storage status of each of the plurality of write buffers;

initialize, in the first scoreboard, the first plurality of entries based on the first storage status;

for each write of a first modified record from a corresponding write buffer of the plurality of write buffers to a home node, clear a first entry of the first plurality of entries based on the first entry corresponding to the corresponding write buffer; and

save a first checkpoint when each entry of the first plurality of entries in the first scoreboard is cleared.

7. The system of claim 6, wherein the key memory is further configured to store:

auxiliary data for storing a description of the resource accessible based on the transmitted key, wherein the resource comprises the record.

8. The system of claim 6, further comprising:

a resource manager configured to;

receive a new key request to generate a new key;

generate a new pseudorandom portion of the new key;

generate a second location identifier based on a free location in the key memory for storing the new pseudorandom portion;

combine the second location identifier with the new pseudorandom portion to obtain the new key;

store the new pseudorandom portion in a second location specified by the second location identifier; and

provide the new key to a requester.

9. The system of claim 6, further comprising:

a storage system for storing a plurality of software instructions of the resource manager.

10. The system of claim 6, further comprising:

a central processing unit; and

a storage system for storing a plurality of records,

wherein providing access to the record identified by the record identifier comprises obtaining the record directly from the storage system while bypassing the central processing unit.

11. The system of claim 6, further comprising:

a second scoreboard comprising a second plurality of entries, wherein each entry of the second plurality of entries corresponds to a write buffer of the plurality of write buffers; and

a second thread configured to;

after the initializing of the first plurality of entries and after a subset of the first plurality of entries is cleared;

reuse at least one of the plurality of write buffers for storing a second modified record without modifying the first plurality of entries based on the storing of the second modified record;

obtain a second storage status of each of the plurality of write buffers;

initialize, in the second scoreboard, the second plurality of entries based on the second storage status;

for each write of a second modified record from the corresponding write buffer of the plurality of write buffers to the home node, clearing a second entry of the second plurality of entries based on the second entry corresponding to the corresponding write buffer; and

saving a second checkpoint when each entry of the second plurality of entries is cleared.

12. A distributed system for record access, comprising:

a requester node configured to transmit a request for a record, wherein the request comprises a transmitted key and a record identifier, the requester node comprising;

a storage system for storing a plurality of write buffers;

a first scoreboard comprising a first plurality of entries, wherein each entry of the first plurality of entries corresponds to a write buffer of the plurality of write buffers; and

a first thread configured to;

obtain a first storage status of each of the plurality of write buffers;

initialize, in the first scoreboard, the first plurality of entries based on the first storage status;

for each write of a first modified record from a corresponding write buffer of the plurality of write buffers to the home node, clear a first entry of the first plurality of entries based on the first entry corresponding to the corresponding write buffer; and

save a first checkpoint when each entry of the first plurality of entries in the first scoreboard is cleared; and

a home node comprising;

an authentication module comprising;

key memory configured to store a plurality of stored pseudorandom portions of a plurality of keys;

a hardware gatekeeper configured to;

receive the request for the record from the requester node;

extract a first location identifier and a transmitted pseudorandom portion from the transmitted key, wherein the location identifier specifies a first location, in the key memory, of a stored pseudorandom portion of the plurality of stored pseudorandom portions;

obtain the stored pseudorandom portion from the first location in the key memory specified by the first location identifier; and

provide access to the record identified by the record identifier by determining that the transmitted pseudorandom portion matches the stored pseudorandom portion.

13. The distributed system of claim 12, wherein the home node further comprises:

a resource manager configured to;

receive a new key request for generating a new key;

generate a new pseudorandom portion of the new key;

generate a second location identifier based on a free location in the key memory for storing the new pseudorandom portion;

combine the second location identifier with the new pseudorandom portion to obtain the new key;

store the new pseudorandom portion in a second location specified by the second location identifier; and

provide the new key to a requester.

14. The distributed system of claim 12, wherein the home node further comprises:

a storage system for storing a plurality of software instructions of the resource manager.

15. The distributed system of claim 12, wherein the home node further comprises:

a central processing unit; and

a storage system for storing a plurality of records,

wherein providing access to a record identified by the record identifier comprises obtaining the record directly from the storage system while bypassing the central processing unit.

16. The distributed system of claim 12, wherein the requester node further comprises:

a second scoreboard comprising a second plurality of entries, wherein each entry of the second plurality of entries corresponds to a write buffer of the plurality of write buffers; and

a second thread configured to;

after the initializing of the first plurality of entries and after a subset of the first plurality of entries is cleared;

reuse at least one of the plurality of write buffers for storing a second modified record without modifying the first plurality of entries based on the storing of the second modified record;

obtain a second storage status of each of the plurality of write buffers;

initialize, in the second scoreboard, the second plurality of entries based on the second storage status;

for each write of a second modified record from the corresponding write buffer of the plurality of write buffers to the home node, clearing a second entry of the second plurality of entries based on the second entry corresponding to the corresponding write buffer; and

saving a second checkpoint when each entry of the second plurality of entries is cleared.

17. The distributed system of claim 16, wherein the first scoreboard corresponds to a first plurality of threads comprising the first thread, and wherein the second scoreboard corresponds to a second plurality of threads comprising the second thread.