Perform batches of RPC requests.
This op asynchronously performs either a single RPC request, or a batch of requests. RPC requests are defined by three main parameters:
- `address` (the host+port or BNS address of the request) - `method` (the RPC method name for the request) - `request` (the serialized proto string, or vector of strings, of the RPC request argument).
For example, if you have an RPC service running on port localhost:2345, and its interface is configured with the following proto declaration:
service MyService {
rpc MyMethod(MyRequestProto) returns (MyResponseProto) {
};
}
address = "localhost:2345"
method = "MyService/MyMethod"
For example, to send a single, empty, `MyRequestProto`, call this op with `request = ""`. To send 5 parallel empty requests, call this op with `request = ["", "", "", "", ""]`.
More generally, one can create a batch of `MyRequestProto` serialized protos from regular batched tensors using the `encode_proto` op, and convert the response `MyResponseProto` serialized protos to batched tensors using the `decode_proto` op.
NOTE Working with serialized proto strings is faster than instantiating actual proto objects in memory, so no performance degradation is expected compared to writing custom kernels for this workflow.
If the connection fails or the remote worker returns an error status, the op reraises this exception locally.
See the `TryRpc` op if you prefer to handle RPC failures manually in the graph.
Nested Classes
class | Rpc.Options | Optional attributes for Rpc
|
Constants
String | OP_NAME | The name of this op, as known by TensorFlow core engine |
Public Methods
Output<TString> |
asOutput()
Returns the symbolic handle of the tensor.
|
static Rpc | |
static Rpc.Options |
failFast(Boolean failFast)
|
static Rpc.Options |
protocol(String protocol)
|
Output<TString> |
response()
Same shape as `request`.
|
static Rpc.Options |
timeoutInMs(Long timeoutInMs)
|
Inherited Methods
Constants
public static final String OP_NAME
The name of this op, as known by TensorFlow core engine
Public Methods
public Output<TString> asOutput ()
Returns the symbolic handle of the tensor.
Inputs to TensorFlow operations are outputs of another TensorFlow operation. This method is used to obtain a symbolic handle that represents the computation of the input.
public static Rpc create (Scope scope, Operand<TString> address, Operand<TString> method, Operand<TString> request, Options... options)
Factory method to create a class wrapping a new Rpc operation.
Parameters
scope | current scope |
---|---|
address | `0-D` or `1-D`. The address (i.e. host_name:port) of the RPC server. If this tensor has more than 1 element, then multiple parallel rpc requests are sent. This argument broadcasts with `method` and `request`. |
method | `0-D` or `1-D`. The method address on the RPC server. If this tensor has more than 1 element, then multiple parallel rpc requests are sent. This argument broadcasts with `address` and `request`. |
request | `0-D` or `1-D`. Serialized proto strings: the rpc request argument. If this tensor has more than 1 element, then multiple parallel rpc requests are sent. This argument broadcasts with `address` and `method`. |
options | carries optional attributes values |
Returns
- a new instance of Rpc
public static Rpc.Options failFast (Boolean failFast)
Parameters
failFast | `boolean`. If `true` (default), then failures to connect (i.e., the server does not immediately respond) cause an RPC failure. |
---|
public static Rpc.Options protocol (String protocol)
Parameters
protocol | RPC protocol to use. Empty string means use the default protocol. Options include 'grpc'. |
---|
public Output<TString> response ()
Same shape as `request`. Serialized proto strings: the rpc responses.
public static Rpc.Options timeoutInMs (Long timeoutInMs)
Parameters
timeoutInMs | `int`. If `0` (default), then the kernel will run the RPC request and only time out if the RPC deadline passes or the session times out. If this value is greater than `0`, then the op will raise an exception if the RPC takes longer than `timeout_in_ms`. |
---|