Microsoft Technology Licensing, LLC

Linkedin Corporation

Andrew Joseph Carter

Scott M Meyer

Andrew Rodriguez

Siddharth Rajendra Shah

Srinath Shankar

SungJu Cho

Joshua D Ehrlich

Ying Lu

Jane Alam Jan

Walaa Eldin M Moustafa

Robert W Stephenson

Karan Rajen Parikh

Yongling Song

Yiming Yang

Sanjay Sachdev

Qingpeng Niu

Roman A Averbukh

Maverick Shu-Bun Lee

Shyam Shankar

Eric Manuel

Yanwei Zhang

Steven Callister

Jiajun Yao

Tiance Liang

Dhruvesh Rathore

The disclosed embodiments provide a system for processing queries of a graph database. During operation, the system stores intermediate results of the query in a structure comprising rows in a set of tables and links between pairs of rows in different tables in the set of tables. Next, the system tracks, in the structure, representations of data, relationships, and path consistency in the query. The system then applies one or more operations to existing rows in the structure to generate a final result of the query. Finally, the system provides the final result in a response to the query.

Enforcing path consistency in graph database path query evaluation

The disclosed technologies are capable of selectively using data structure operations for path query evaluation. One technique involves reading a query that traverses at least two nodes and at least one edge of a graph in a graph database; compiling the query into a set of variables and a set of constraints, where the set of variables and the set of constraints correspond to the two nodes and the one edge of the graph; creating an in-memory data structure that comprises a table; using the set of variables and the set of constraints to determine an operation that is performable using the in-memory data structure; checking for an existence of a condition relating to the in-memory data structure or the operation; skipping the operation if the condition exists or executing the operation if the condition does not exist; and storing a set of intermediate query results in the table.

Selective use of data structure operations for path query evaluation

The disclosed embodiments provide a system for processing queries of a graph database storing a graph, wherein the graph comprises a set of edges defined by a first linkage, a second linkage, and a third linkage. During operation, the system maintains the base version of an index of the graph database. Upon branching a version of the graph database from a first offset representing a virtual time in the base version of the graph database, the system creates a branched version of the index from a second offset corresponding to the virtual time in the base version of the index. The system then processes queries of the graph database based on the offsets and references from the branched version of the index to the base version of the index.

Branch threading in graph databases

The disclosed embodiments provide a system for processing queries of a graph database. During operation, the system stores intermediate results of a query in a structure comprising rows in a set of tables. Next, the system applies one or more operations to existing rows in the structure to generate additional rows in additional tables of the structure. The system also tracks inputs and outputs associated with the one or more operations in a set of relationships between rows of different tables in the structure. Finally, the system uses the rows and the set of relationships to generate a final result of the query and provides the final result in a response to the query.

Path query evaluation in graph databases

The disclosed embodiments provide a system for managing a materialized view of data in a data store. During operation, the system obtains a first creation time of a first materialized view of the data in the data store. Next, the system verifies that the first materialized view is not affected by deletions applied to the data since the first creation time. The system then identifies one or more additions to the data that are not included in the first materialized view. Finally, the system applies the additions to the first materialized view to maintain an eventual consistency of the first materialized view with the data store.

Managing materialized views in eventually consistent distributed data stores

During operation, a system for processing queries of a graph database executes a set of processes for processing queries of a graph database storing a graph, wherein the graph comprises a set of nodes, a set of edges between pairs of nodes in the set of nodes, and a set of predicates. When a query of the graph database is received, the system obtains, from the query, a compound comprising a compound type and a set of identity-giving nodes. Next, the system uses the compound to access a tuple for storage in a log-based representation of the graph database, wherein the tuple includes the compound type and the identity-giving nodes. The system then uses the tuple to generate a result of the query and provides the result in a response to the query.

Supporting tuples in log-based representations of graph databases

The disclosed embodiments provide a system for processing queries of a graph database. During operation, the system executes a set of processes for processing queries of a graph database storing a graph, wherein the graph comprises a set of nodes, a set of edges between pairs of nodes in the set of nodes, and a set of predicates. When a query of the graph database is received, the system performs a lookup of a hash map to obtain one or more offsets into a compound store for the graph database, which includes identity-giving nodes for a set of tuples in the graph database. Next, the system accesses the offset(s) in the compound store to obtain a subset of tuples matching the query. The system then uses the subset of tuples to generate a result of the query and provides the result in a response to the query.

Compound indexes for graph databases

The disclosed embodiments provide a system for processing data. During operation, the system determines a current incoming queries per second (QPS) to one or more components for processing queries of a graph database, wherein the graph database is replicated across multiple clusters and distributed among a set of storage nodes in each of the clusters. Next, the system uses the current incoming QPS to estimate, for the one or more components, an expected QPS associated with fanning out of the queries to the clusters. The system then selects a number of clusters in the multiple clusters for fanning out of a query based on the expected QPS and one or more throughput limits for the one or more components. Finally, the system transmits the query to one or more of the storage nodes in the selected number of clusters.

Throughput-based fan-out control in scalable distributed data stores

A system executes a query associated with an application against a graph database by providing, to a first shard of the graph database, the query and a first query header that specifies the first shard. The query includes a subject, a predicate and an object, and the graph database stores a graph that includes nodes, edges between the nodes, and predicates to represent and store data. In response to the query, the system receives results and associated result headers from the first shard and a second shard, where the result headers specify that the results are partial results that are particular fractions of a total result. Furthermore, a combination of the partial results provides the total result to the query that includes a subset of the graph. Note that the subset of the graph may include desired information expressed within an associated structure of the graph.

Message passing in a distributed graph database

A social graph is divided into a series of partial graphs having limited hops and reference counts. For each of a plurality of nodes in the social graph, a partial graph for the node is created having a first degree list of nodes reachable in one hop and a second degree list of nodes reachable in two hops. Reference counts of how many paths exist between the node and each node reachable in two hops are also added to the second degree list. A global inverted index is maintained containing a list of all nodes in the social graph and for each node in the social graph a list of nodes that directly connect to the node. The partial graphs created for each of the plurality of nodes are distributed across a plurality of data servers. An index of which partial graphs are stored on which data servers is maintained.

Partial graph incremental update in a social network

Techniques are provided for minimizing latency due to garbage collection in a distributed system. Time data is stored at a computing device. The time data indicates when each of multiple computing nodes (e.g., servers) in the distributed system might be performing (or is scheduled to perform) a garbage collection operation. The computing device receives a first request and, in response, determines, based on the time data, one or more computing nodes to process a second (e.g., related) request, which may contain identical content as in the first request. The second request is sent to the one or more computing nodes.

Minimizing latency due to garbage collection in a distributed system

In order to request desired information from a graph database, a system executes a query against the graph database. The graph database may store a graph that includes nodes, edges between the nodes, and predicates to represent and store data with index-free adjacency. Moreover, the query may identify a first edge associated with a predicate that specifies one or more of the nodes in the graph. In response to the query, the system receives a result that includes a subset of the graph. In particular, the subset of the graph may include the desired information expressed within an associated structure of the graph.

Graph-based queries

In order to request desired information from a graph database, a system executes a query against the graph database. The graph database may store a graph that includes nodes, edges between the nodes, and predicates to represent and store data with index-free adjacency. Moreover, the query may identify a first edge associated with a predicate that specifies one or more of the nodes in the graph. In response to the query, the system receives a result that includes a subset of the graph. Then, the system may use the result of the query in a concatenated set of queries. In particular, the system may execute a second query against the subset of the graph. Furthermore, in response to the second query, the system may receive a second result that includes a second subset of the graph.

Growing community of inventors

Mountain View, CA, United States of America

Andrew Joseph Carter