Typedb-protocol

Documentation: http://docs.vaticle.com/docs/client-api/java

Distribution

For Java through Maven

Available through https://repo.vaticle.com
xml
<repositories>
<repository>
<id>repo.vaticle.com</id>
<url>https://repo.vaticle.com/repository/maven/</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupid>com.vaticle.typedb</groupid>
<artifactid>typedb-protocol</artifactid>
<version>2.1.0</version>
</dependency>
</dependencies>



New Features


Bugs Fixed


Code Refactors


Other Improvements

Documentation: http://dev.grakn.ai/docs/client-api/java

Distribution

For Java through Maven

Available through https://repo.grakn.ai
xml
<repositories>
<repository>
<id>repo.grakn.ai</id>
<url>https://repo.grakn.ai/repository/maven/</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-session</artifactid>
<version>2.0.1</version>
</dependency>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-keyspace</artifactid>
<version>2.0.1</version>
</dependency>
</dependencies>



New Features

- **Store request IDs as bytes; add 'inferred' to Thing**
Previously the field `req_id` of a transaction request was stored as a string, which is an inefficient representation of a UUID. We've changed it to bytes.
Also, we deleted `IsInferred` and added the `inferred` flag to `Thing`, allowing clients to check if a Thing is inferred without an additional network roundtrip.


Bugs Fixed


Code Refactors


Other Improvements

Documentation: http://dev.grakn.ai/docs/client-api/java

Distribution

For Java through Maven

Available through https://repo.grakn.ai
xml
<repositories>
<repository>
<id>repo.grakn.ai</id>
<url>https://repo.grakn.ai/repository/maven/</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-session</artifactid>
<version>2.0.0</version>
</dependency>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-keyspace</artifactid>
<version>2.0.0</version>
</dependency>
</dependencies>



New Features

- **Implement Explanations in RPC**
We add Explanations, and the explain() query endpoint to the protocol. This enables all clients to retrieve all available explanations for a particular explainable concept map.

- **Introduced transaction request batching**
When sending large amounts of requests from the client to server, it is possible to reach GRPC's maximum throughput of messages per second, on large scale systems. For example, a client with 32 cores, can have 1000 concurrent transactions sending requests to the server. This will easily go beyond what GRPC can handle, and thus becomes the bottleneck in the system. That client will not be sending as many requests it wants per second, and the server may still be able to handle more requests to handle. We have now overcome this problem, by allowing the transaction stream (the predominant GRPC service) to _"batch"_ the transaction requests into one GRPC message to send to the server, within a given time window. In the client-side implementations, we have chosen the time window to be 1 millisecond - small enough to not cause any noticeable performance degradation on individual requests, but big enough to _"batch"_ all requests within that time window into one message and multiply the throughput.
Additionally, we also refined the messages to control iterators - `continue` and `done` - to be replaced with `Transaction.Iterate.Req` and `Transaction.Iterate.Res`, and we moved `latency_millis` into `Transaction.Open.Req`.


Bugs Fixed


Code Refactors


Other Improvements

Documentation: http://dev.grakn.ai/docs/client-api/java

Distribution

For Java through Maven

Available through https://repo.grakn.ai
xml
<repositories>
<repository>
<id>repo.grakn.ai</id>
<url>https://repo.grakn.ai/repository/maven/</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-session</artifactid>
<version>2.0.0-alpha-12</version>
</dependency>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-keyspace</artifactid>
<version>2.0.0-alpha-12</version>
</dependency>
</dependencies>



New Features

- **Add 'type' field to Thing**
The `Thing` message now contains a new field, `type`, which holds the concept type. As a result of this change, `value_type` is now redundant (equivalent to `type.value_type`) and `encoding` can be inferred from `type.encoding`, so both fields have been deleted.


Bugs Fixed


Code Refactors


Other Improvements

Documentation: http://dev.grakn.ai/docs/client-api/java

Distribution

For Java through Maven

Available through https://repo.grakn.ai
xml
<repositories>
<repository>
<id>repo.grakn.ai</id>
<url>https://repo.grakn.ai/repository/maven/</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-session</artifactid>
<version>2.0.0-alpha-11</version>
</dependency>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-keyspace</artifactid>
<version>2.0.0-alpha-11</version>
</dependency>
</dependencies>



New Features

- **Add parallel option to Options**
To allow clients to disable parallel execution, we add the `parallel_opt` to `Options.


Bugs Fixed


Code Refactors


Other Improvements

Documentation: http://dev.grakn.ai/docs/client-api/java

Distribution

For Java through Maven

Available through https://repo.grakn.ai
xml
<repositories>
<repository>
<id>repo.grakn.ai</id>
<url>https://repo.grakn.ai/repository/maven/</url>
</repository>
</repositories>
<dependencies>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-session</artifactid>
<version>2.0.0-alpha-10</version>
</dependency>
<dependency>
<groupid>io.grakn.protocol</groupid>
<artifactid>grakn-keyspace</artifactid>
<version>2.0.0-alpha-10</version>
</dependency>
</dependencies>



New Features

- **Use self-hosted Bazel cache**
Speed up CI by using self-hosted remote Bazel cache

- **Add inference logging option**
Add an option that allows users to turn on or off resolution tracing for reasoner.

- **Add methods to get and list Cluster databases**
We added new service methods to Cluster that retrieve Cluster databases, including their replica info. Although this was possible before with `database_replicas`, this method was not well aligned with what Core does. Using the terms `get` and `all` allows us to establish a more Core-like API.


Bugs Fixed


Code Refactors


Other Improvements