De-serialized datastore transaction system
AssigneeSubstrate Inc.
InventorsChristian Beaumont, Behnaz Beaumont, Jouke van der Maas, Jan Drake
Life of this patent
Estimating the term…
Abstract
A de-serialized data store and transaction system includes a content addressable memory mapped to an n-tuple-based multimodal data structure, the n-tuple-based multimodal data structure supporting heterogeneous nested data types including a streamable abstract syntax tree and a self-describing operational interface comprising commands transmitted via a plurality of nested s-expressions.
Claims38
A method of controlling read and write access to a memory structure comprising:
- initiating a read lock comprising:
- obtaining a reader pool ID for a thread from a fixed pool of readers;
- waiting for a writer to finish by entering a wait-loop and querying a scheduler to reschedule the thread if current wait time exceeds a threshold value;
- declaring a resource to be read;
- checking for an active write lock;
- and
- returning the reader pool ID for the thread;
- and
- initiating a write-lock comprising:
- checking for an active write lock flag and an active read lock flag and entering a wait-loop if the active write lock flag or the active read lock flag is present;
- and
- querying a scheduler to reschedule the thread if the wait time exceeds the threshold value.
A de-serialized data store and transaction system comprising:
- a content addressable memory mapped to an n-tuple-based multimodal data structure, the n-tuple-based multimodal data structure supporting heterogeneous nested data types comprising:
- a streamable abstract syntax tree;
- and
- a self-describing operational interface comprising commands transmitted via a plurality of nested s-expressions.
A secure proxy-free data store access system comprising:
- A plurality of hierarchically privileged nested tuple-space partitions in a content addressable memory;
- a plurality of hierarchically contained programming interface functions defined within each of the plurality of hierarchically privileged nested tuple-space partitions;
- a plurality of virtual machines each associated with a processor core associated with at least one tuple-space partition;
- and
- a reading and writing data from the content addressable memory via a transactional read pipeline and a transactional write pipeline.
A unified data store and transaction system comprising:
- a querying an n-tuple-based multimodal data structure via a mutable tuple-based interface the mutable tuple-based interface comprising:
- a memory controller;
- and
- a query operation set;
- receiving a tuple from a mutable tuple-based query interface with a tuple-reader and reading the tuple into a tuple object;
- and
- evaluating the tuple object against semantic rules via a tuple evaluator.
A taxonomy model in a unified data store comprising:
- a data graph structure comprising a plurality of self-describing data objects each self-describing data object comprising a plurality of:
- a unique identifier;
- a relationship object linking the self-describing data object to a plurality of other self-describing data objects;
- a domain object linking the self-describing data object to a plurality of domains;
- and
- a token object representing the self-describing data object within a domain;
- an n-tuple-based multimodal data structure a content addressable memory;
- a tuple-reader;
- and
- a tuple evaluator configured with a set of tuple semantic rules.
A method of implementing a multimodal split-associative data store and control comprising:
- receiving data in the form of a plurality of n-tuples via a transactional pipeline interface;
- transmitting the data to an in-memory multimodal data processor within a content-addressable memory, the in-memory multimodal data processor comprising an associative n-tuple store;
- and
- receiving a plurality of commands to execute a plurality of actions on the data in the associative n-tuple store.
Dependencies are inferred from the claim text.