Arrow Flight RPC

Arrow Flight is an RPC framework for high-performance data services based on Arrow data, and is built on top of gRPC and the IPC format.

Flight is organized around streams of Arrow record batches, being either downloaded from or uploaded to another service. A set of metadata methods offers discovery and introspection of streams, as well as the ability to implement application-specific methods.

Methods and message wire formats are defined by Protobuf, enabling interoperability with clients that may support gRPC and Arrow separately, but not Flight. However, Flight implementations include further optimizations to avoid overhead in usage of Protobuf (mostly around avoiding excessive memory copies).

RPC Methods and Request Patterns

Flight defines a set of RPC methods for uploading/downloading data, retrieving metadata about a data stream, listing available data streams, and for implementing application-specific RPC methods. A Flight service implements some subset of these methods, while a Flight client can call any of these methods.

Data streams are identified by descriptors (the FlightDescriptor message), which are either a path or an arbitrary binary command. For instance, the descriptor may encode a SQL query, a path to a file on a distributed file system, or even a pickled Python object; the application can use this message as it sees fit.

Thus, one Flight client can connect to any service and perform basic operations. To facilitate this, Flight services are expected to support some common request patterns, described next. Of course, applications may ignore compatibility and simply treat the Flight RPC methods as low-level building blocks for their own purposes.

See Protocol Buffer Definitions for full details on the methods and messages involved.

Downloading Data

A client that wishes to download the data would:

Retrieving data via DoGet.

1. Construct or acquire a FlightDescriptor for the data set they are interested in.

   A client may know what descriptor they want already, or they may use methods like ListFlights to discover them.

2. Call GetFlightInfo(FlightDescriptor) to get a FlightInfo message.

   Flight does not require that data live on the same server as metadata. Hence, FlightInfo contains details on where the data is located, so the client can go fetch the data from an appropriate server. This is encoded as a series of FlightEndpoint messages inside FlightInfo. Each endpoint represents some location that contains a subset of the response data.

   An endpoint contains a list of locations (server addresses) where this data can be retrieved from, and a Ticket, an opaque binary token that the server will use to identify the data being requested. There is no ordering defined on endpoints or the data within, so if the dataset is sorted, applications should return data in a single endpoint.

   The response also contains other metadata, like the schema, and optionally an estimate of the dataset size.

3. Consume each endpoint returned by the server.

   To consume an endpoint, the client should connect to one of the locations in the endpoint, then call DoGet(Ticket) with the ticket in the endpoint. This will give the client a stream of Arrow record batches.

   If the server wishes to indicate that the data is on the local server and not a different location, then it can return an empty list of locations. The client can then reuse the existing connection to the original server to fetch data. Otherwise, the client must connect to one of the indicated locations.

   In this way, the locations inside an endpoint can also be thought of as performing look-aside load balancing or service discovery functions. And the endpoints can represent data that is partitioned or otherwise distributed.

   The client must consume all endpoints to retrieve the complete data set. The client can consume endpoints in any order, or even in parallel, or distribute the endpoints among multiple machines for consumption; this is up to the application to implement.

Uploading Data

To upload data, a client would:

Uploading data via DoGet.

1. Construct or acquire a FlightDescriptor, as before.

2. Call DoPut(FlightData) and upload a stream of Arrow record batches.

   The FlightDescriptor is included with the first message so the server can identify the dataset.

DoPut allows the server to send response messages back to the client with custom metadata. This can be used to implement things like resumable writes (e.g. the server can periodically send a message indicating how many rows have been committed so far).

Exchanging Data

Some use cases may require uploading and downloading data within a single call. While this can be emulated with multiple calls, this may be difficult if the application is stateful. For instance, the application may wish to implement a call where the client uploads data and the server responds with a transformation of that data; this would require being stateful if implemented using DoGet and DoPut. Instead, DoExchange allows this to be implemented as a single call. A client would:

Complex data flow with DoExchange.

1. Construct or acquire a FlightDescriptor, as before.

2. Call DoExchange(FlightData).

   The FlightDescriptor is included with the first message, as with DoPut. At this point, both the client and the server may simultaneously stream data to the other side.

Authentication

Flight supports a variety of authentication methods that applications can customize for their needs.

“Handshake” authentication

This is implemented in two parts. At connection time, the client calls the Handshake RPC method, and the application-defined authentication handler can exchange any number of messages with its counterpart on the server. The handler then provides a binary token. The Flight client will then include this token in the headers of all future calls, which is validated by the server authentication handler.

Applications may use any part of this; for instance, they may ignore the initial handshake and send an externally acquired token (e.g. a bearer token) on each call, or they may establish trust during the handshake and not validate a token for each call, treating the connection as stateful (a “login” pattern).

Warning

Unless a token is validated on every call, this pattern is not secure, especially in the presenence of a layer 7 load balancer, as is common with gRPC, or if gRPC transparently reconnects the client.

Header-based/middleware-based authentication

Clients may include custom headers with calls. Custom middleware can then be implemented to validate and accept/reject calls on the server side.

Mutual TLS (mTLS)

The client provides a certificate during connection establishment which is verified by the server. The application does not need to implement any authentication code, but must provision and distribute certificates.

This may only be available in certain implementations, and is only available when TLS is also enabled.

Some Flight implementations may expose the underlying gRPC API as well, in which case any authentication method supported by gRPC is available.

Transport Implementations

Flight is primarily defined in terms of its Protobuf and gRPC specification below, but Arrow implementations may also support alternative transports (see Flight RPC). In that case, implementations should use the following URI schemes for the given transport implemenatations:

Transport	URI Scheme
gRPC (plaintext)	grpc: or grpc+tcp:
gRPC (TLS)	grpc+tls:
gRPC (Unix domain socket)	grpc+unix:
UCX (plaintext)	ucx:

Error Handling

Arrow Flight defines its own set of error codes. The implementation differs between languages (e.g. in C++, Unimplemented is a general Arrow error status while it’s a Flight-specific exception in Java), but the following set is exposed:

Error Code	Description
UNKNOWN	An unknown error. The default if no other error applies.
INTERNAL	An error internal to the service implementation occurred.
INVALID_ARGUMENT	The client passed an invalid argument to the RPC.
TIMED_OUT	The operation exceeded a timeout or deadline.
NOT_FOUND	The requested resource (action, data stream) was not found.
ALREADY_EXISTS	The resource already exists.
CANCELLED	The operation was cancelled (either by the client or the server).
UNAUTHENTICATED	The client is not authenticated.
UNAUTHORIZED	The client is authenticated, but does not have permissions for the requested operation.
UNIMPLEMENTED	The RPC is not implemented.
UNAVAILABLE	The server is not available. May be emitted by the client for connectivity reasons.

External Resources

• https://arrow.apache.org/blog/2019/10/13/introducing-arrow-flight/

• https://arrow.apache.org/blog/2018/10/09/0.11.0-release/

• https://www.slideshare.net/JacquesNadeau5/apache-arrow-flight-overview

Protocol Buffer Definitions

/*
 * Licensed to the Apache Software Foundation (ASF) under one
 * or more contributor license agreements.  See the NOTICE file
 * distributed with this work for additional information
 * regarding copyright ownership.  The ASF licenses this file
 * to you under the Apache License, Version 2.0 (the
 * "License"); you may not use this file except in compliance
 * with the License.  You may obtain a copy of the License at
 * <p>
 * http://www.apache.org/licenses/LICENSE-2.0
 * <p>
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

syntax = "proto3";

option java_package = "org.apache.arrow.flight.impl";
option go_package = "github.com/apache/arrow/go/flight;flight";
option csharp_namespace = "Apache.Arrow.Flight.Protocol";

package arrow.flight.protocol;

/*
 * A flight service is an endpoint for retrieving or storing Arrow data. A
 * flight service can expose one or more predefined endpoints that can be
 * accessed using the Arrow Flight Protocol. Additionally, a flight service
 * can expose a set of actions that are available.
 */
service FlightService {

  /*
   * Handshake between client and server. Depending on the server, the
   * handshake may be required to determine the token that should be used for
   * future operations. Both request and response are streams to allow multiple
   * round-trips depending on auth mechanism.
   */
  rpc Handshake(stream HandshakeRequest) returns (stream HandshakeResponse) {}

  /*
   * Get a list of available streams given a particular criteria. Most flight
   * services will expose one or more streams that are readily available for
   * retrieval. This api allows listing the streams available for
   * consumption. A user can also provide a criteria. The criteria can limit
   * the subset of streams that can be listed via this interface. Each flight
   * service allows its own definition of how to consume criteria.
   */
  rpc ListFlights(Criteria) returns (stream FlightInfo) {}

  /*
   * For a given FlightDescriptor, get information about how the flight can be
   * consumed. This is a useful interface if the consumer of the interface
   * already can identify the specific flight to consume. This interface can
   * also allow a consumer to generate a flight stream through a specified
   * descriptor. For example, a flight descriptor might be something that
   * includes a SQL statement or a Pickled Python operation that will be
   * executed. In those cases, the descriptor will not be previously available
   * within the list of available streams provided by ListFlights but will be
   * available for consumption for the duration defined by the specific flight
   * service.
   */
  rpc GetFlightInfo(FlightDescriptor) returns (FlightInfo) {}

  /*
   * For a given FlightDescriptor, get the Schema as described in Schema.fbs::Schema
   * This is used when a consumer needs the Schema of flight stream. Similar to
   * GetFlightInfo this interface may generate a new flight that was not previously
   * available in ListFlights.
   */
   rpc GetSchema(FlightDescriptor) returns (SchemaResult) {}

  /*
   * Retrieve a single stream associated with a particular descriptor
   * associated with the referenced ticket. A Flight can be composed of one or
   * more streams where each stream can be retrieved using a separate opaque
   * ticket that the flight service uses for managing a collection of streams.
   */
  rpc DoGet(Ticket) returns (stream FlightData) {}

  /*
   * Push a stream to the flight service associated with a particular
   * flight stream. This allows a client of a flight service to upload a stream
   * of data. Depending on the particular flight service, a client consumer
   * could be allowed to upload a single stream per descriptor or an unlimited
   * number. In the latter, the service might implement a 'seal' action that
   * can be applied to a descriptor once all streams are uploaded.
   */
  rpc DoPut(stream FlightData) returns (stream PutResult) {}

  /*
   * Open a bidirectional data channel for a given descriptor. This
   * allows clients to send and receive arbitrary Arrow data and
   * application-specific metadata in a single logical stream. In
   * contrast to DoGet/DoPut, this is more suited for clients
   * offloading computation (rather than storage) to a Flight service.
   */
  rpc DoExchange(stream FlightData) returns (stream FlightData) {}

  /*
   * Flight services can support an arbitrary number of simple actions in
   * addition to the possible ListFlights, GetFlightInfo, DoGet, DoPut
   * operations that are potentially available. DoAction allows a flight client
   * to do a specific action against a flight service. An action includes
   * opaque request and response objects that are specific to the type action
   * being undertaken.
   */
  rpc DoAction(Action) returns (stream Result) {}

  /*
   * A flight service exposes all of the available action types that it has
   * along with descriptions. This allows different flight consumers to
   * understand the capabilities of the flight service.
   */
  rpc ListActions(Empty) returns (stream ActionType) {}

}

/*
 * The request that a client provides to a server on handshake.
 */
message HandshakeRequest {

  /*
   * A defined protocol version
   */
  uint64 protocol_version = 1;

  /*
   * Arbitrary auth/handshake info.
   */
  bytes payload = 2;
}

message HandshakeResponse {

  /*
   * A defined protocol version
   */
  uint64 protocol_version = 1;

  /*
   * Arbitrary auth/handshake info.
   */
  bytes payload = 2;
}

/*
 * A message for doing simple auth.
 */
message BasicAuth {
  string username = 2;
  string password = 3;
}

message Empty {}

/*
 * Describes an available action, including both the name used for execution
 * along with a short description of the purpose of the action.
 */
message ActionType {
  string type = 1;
  string description = 2;
}

/*
 * A service specific expression that can be used to return a limited set
 * of available Arrow Flight streams.
 */
message Criteria {
  bytes expression = 1;
}

/*
 * An opaque action specific for the service.
 */
message Action {
  string type = 1;
  bytes body = 2;
}

/*
 * An opaque result returned after executing an action.
 */
message Result {
  bytes body = 1;
}

/*
 * Wrap the result of a getSchema call
 */
message SchemaResult {
  // The schema of the dataset in its IPC form:
  //   4 bytes - an optional IPC_CONTINUATION_TOKEN prefix
  //   4 bytes - the byte length of the payload
  //   a flatbuffer Message whose header is the Schema
  bytes schema = 1;
}

/*
 * The name or tag for a Flight. May be used as a way to retrieve or generate
 * a flight or be used to expose a set of previously defined flights.
 */
message FlightDescriptor {

  /*
   * Describes what type of descriptor is defined.
   */
  enum DescriptorType {

    // Protobuf pattern, not used.
    UNKNOWN = 0;

    /*
     * A named path that identifies a dataset. A path is composed of a string
     * or list of strings describing a particular dataset. This is conceptually
     *  similar to a path inside a filesystem.
     */
    PATH = 1;

    /*
     * An opaque command to generate a dataset.
     */
    CMD = 2;
  }

  DescriptorType type = 1;

  /*
   * Opaque value used to express a command. Should only be defined when
   * type = CMD.
   */
  bytes cmd = 2;

  /*
   * List of strings identifying a particular dataset. Should only be defined
   * when type = PATH.
   */
  repeated string path = 3;
}

/*
 * The access coordinates for retrieval of a dataset. With a FlightInfo, a
 * consumer is able to determine how to retrieve a dataset.
 */
message FlightInfo {
  // The schema of the dataset in its IPC form:
  //   4 bytes - an optional IPC_CONTINUATION_TOKEN prefix
  //   4 bytes - the byte length of the payload
  //   a flatbuffer Message whose header is the Schema
  bytes schema = 1;

  /*
   * The descriptor associated with this info.
   */
  FlightDescriptor flight_descriptor = 2;

  /*
   * A list of endpoints associated with the flight. To consume the
   * whole flight, all endpoints (and hence all Tickets) must be
   * consumed. Endpoints can be consumed in any order.
   *
   * In other words, an application can use multiple endpoints to
   * represent partitioned data.
   *
   * There is no ordering defined on endpoints. Hence, if the returned
   * data has an ordering, it should be returned in a single endpoint.
   */
  repeated FlightEndpoint endpoint = 3;

  // Set these to -1 if unknown.
  int64 total_records = 4;
  int64 total_bytes = 5;
}

/*
 * A particular stream or split associated with a flight.
 */
message FlightEndpoint {

  /*
   * Token used to retrieve this stream.
   */
  Ticket ticket = 1;

  /*
   * A list of URIs where this ticket can be redeemed via DoGet().
   *
   * If the list is empty, the expectation is that the ticket can only
   * be redeemed on the current service where the ticket was
   * generated.
   *
   * If the list is not empty, the expectation is that the ticket can
   * be redeemed at any of the locations, and that the data returned
   * will be equivalent. In this case, the ticket may only be redeemed
   * at one of the given locations, and not (necessarily) on the
   * current service.
   *
   * In other words, an application can use multiple locations to
   * represent redundant and/or load balanced services.
   */
  repeated Location location = 2;
}

/*
 * A location where a Flight service will accept retrieval of a particular
 * stream given a ticket.
 */
message Location {
  string uri = 1;
}

/*
 * An opaque identifier that the service can use to retrieve a particular
 * portion of a stream.
 *
 * Tickets are meant to be single use. It is an error/application-defined
 * behavior to reuse a ticket.
 */
message Ticket {
  bytes ticket = 1;
}

/*
 * A batch of Arrow data as part of a stream of batches.
 */
message FlightData {

  /*
   * The descriptor of the data. This is only relevant when a client is
   * starting a new DoPut stream.
   */
  FlightDescriptor flight_descriptor = 1;

  /*
   * Header for message data as described in Message.fbs::Message.
   */
  bytes data_header = 2;

  /*
   * Application-defined metadata.
   */
  bytes app_metadata = 3;

  /*
   * The actual batch of Arrow data. Preferably handled with minimal-copies
   * coming last in the definition to help with sidecar patterns (it is
   * expected that some implementations will fetch this field off the wire
   * with specialized code to avoid extra memory copies).
   */
  bytes data_body = 1000;
}

/**
 * The response message associated with the submission of a DoPut.
 */
message PutResult {
  bytes app_metadata = 1;
}

previous

Arrow Columnar Format

next

Arrow Flight SQL