Systems and methods for preventing data collisions in multiple access postal system data storage systems
View Patent ↗An illustrative data structure is provided using a structured table that uses a split primary key including a first key, referred to as a partition key, and a second key, referred to as a row key, along with a timestamp key, but that does not otherwise utilize uniform columns that can be used as a key. The partition key that is utilized is unusually fine in that it consists of an individual user identifier that is unique to the system and/or across multiple systems used in an enterprise. Additionally, the system can limit access to one session per individual partition key.
1. A non-transitory computer-readable storage medium accessible by a computer program system being executed on a data processing and storage system that has multiple concurrent users, comprising:
a structured data table distributed across at least two nodes including a first node and a second node,
the structured data table comprising at least two non-uniform rows, including at least a first row and a second row,
each of the at least two non-uniform rows including a first key, a second key, and non-uniform columns including a plurality of name/value pair elements,
wherein the first key comprises a partition key that is used in determining which of the at least two nodes stores each of the at least two non-uniform rows such that each of the at least two non-uniform rows having an identical partition key are stored in the same one of the at least two nodes,
wherein the partition key consists of one of a plurality of unique identifiers associated with a respective one of the users, each of the unique identifiers corresponds one-to-one with a corresponding one of the users, and each of the user identifiers is globally unique for the corresponding user across the computer program system, and
wherein the computer program system comprises a postage processing system and each of the users purchases postage using the computer program system, the data processing and storage system is a Cloud Computing system, at least two of the at least two non-uniform rows comprise a different number of columns, at least two of the at least two non-uniform rows comprise a column each having a different type of name/value pair, and the unique identifier associated with one of the users comprises a customer number.
2. The non-transitory computer-readable storage medium of claim 1 , wherein the second key comprises a Row key, and wherein each Row key is globally unique for each of the at least two non-uniform rows.
3. The non-transitory computer-readable storage medium of claim 1 , wherein the second key comprises a Row key, and wherein each Row key is locally unique for each of the at least two non-uniform rows that have the same partition key.
4. The non-transitory computer-readable storage medium of claim 1 , wherein each of the unique identifiers corresponds one-to-one with a client computing device that is used to access the computer programming system by at least one of the users.
5. A method for controlling user session access to a computer program system being executed on a data processing and storage system that has multiple concurrent users and that has a structured data table distributed across at least two nodes including a first node and a second node, the structured data table comprising at least two non-uniform rows, including at least a first row and a second row, each of the at least two non-uniform rows including a first key, a second key, and non-uniform columns including a plurality of name/value pair elements, wherein the first key comprises a partition key that is used in determining which of the at least two nodes stores each of the at least two non-uniform rows such that each of the at least two non-uniform rows having an identical partition key are stored in the same one of the at least two nodes, and wherein the partition key consists of one of a plurality of unique identifiers associated with a respective one of the users, each of the unique identifiers corresponding one-to-one with a corresponding one of the users, and each of the user identifiers is globally unique for the corresponding user across the computer program system, the method comprising:
obtaining a new session request from one user of the multiple concurrent users;
determining the variable type of the partition key;
obtaining a first value of the variable type from the one user;
determining if the first value is listed on an active session log including a plurality of values of the variable type; and
allowing the new session to start with access to the structured table after it is determined that the first value is not listed on the active session log, wherein the computer program system comprises a postage processing system and each of the users purchases postage using the computer program system, the data processing and storage system is a Cloud Computing system, at least two of the at least two non-uniform rows comprise a different number of columns, at least two of the at least two non-uniform rows comprise a column each having a different type of name/value pair, and the unique identifier associated with one of the users comprises a customer number.
6. The method of claim 5 , further comprising the data processing system executing instructions comprising after allowing the new session to start after it is determined that the first value is not listed on the active session log, updating the active session log to include the first value.
7. The method of claim 5 , further comprising:
determining that the new session has ended or timed; and
updating the active session log to remove the first value.