Hi,
I am planning to implement this on IBM PPC64 based systems. I got trouble in starting tensorflow-serving server, I got help from https://github.com/thammegowda/tensorflow-grpc-java and have tensorflow-serving running with inception inference model in place:

(srikanth) [root@brazossrik01 bin]# nohup tensorflow_model_server --model_name=inception --model_base_path=/root/srikanth/SERVING_INCEPTION/SERVING_INCEPTION --port=9000 2>&1 &
[1] 357
(srikanth) [root@brazossrik01 bin]# nohup: ignoring input and appending output to 'nohup.out'

(srikanth) [root@brazossrik01 bin]# ps -eaf |grep 9000
root 357 32464 27 23:03 pts/1 00:00:01 tensorflow_model_server --model_name=inception --model_base_path=/root/srikanth/SERVING_INCEPTION/SERVING_INCEPTION --port=9000
root 574 32464 0 23:03 pts/1 00:00:00 grep --color=auto 9000
(srikanth) [root@brazossrik01 bin]#

However, when I am trying to use the client code, I am getting UNIMPLEMENTED Error when I bind to kafakcat:

(srikanth) [root@brazossrik01 bin]# echo -e "src/main/resources/example.jpg" | /root/srikanth/kafkacat-master/kafkacat -b localhost:9092 -P -t ImageInputTopic
(srikanth) [root@brazossrik01 bin]#

(srikanth) [root@brazossrik01 bin]# java -cp target/tensorflow-serving-java-grpc-kafka-streams-1.0-jar-with-dependencies.jar com.github.megachucky.kafka.streams.machinelearning.Kafka_Streams_TensorFlow_Serving_gRPC_Example
SLF4J: Failed to load class "org.slf4j.impl.StaticLoggerBinder".
SLF4J: Defaulting to no-operation (NOP) logger implementation
SLF4J: See http://www.slf4j.org/codes.html#StaticLoggerBinder for further details.
Image Recognition Microservice is running...
Input images arrive at Kafka topic ImageInputTopic; Output predictions going to Kafka topic ImageOutputTopic
Image path: src/main/resources/example.jpg
Image = src/main/resources/example.jpg
io.grpc.StatusRuntimeException: UNIMPLEMENTED
at io.grpc.stub.ClientCalls.toStatusRuntimeException(ClientCalls.java:222)
at io.grpc.stub.ClientCalls.getUnchecked(ClientCalls.java:203)
at io.grpc.stub.ClientCalls.blockingUnaryCall(ClientCalls.java:132)
at com.github.megachucky.kafka.streams.machinelearning.InceptionBlockingStub.classify(InceptionBlockingStub.java:63)
at com.github.megachucky.kafka.streams.machinelearning.TensorflowObjectRecogniser.recognise(TensorflowObjectRecogniser.java:66)
at com.github.megachucky.kafka.streams.machinelearning.Kafka_Streams_TensorFlow_Serving_gRPC_Example.lambda$main0ドル(Kafka_Streams_TensorFlow_Serving_gRPC_Example.java:91)
at org.apache.kafka.streams.kstream.internals.AbstractStream2ドル.apply(AbstractStream.java:111)
at org.apache.kafka.streams.kstream.internals.KStreamMapValues$KStreamMapProcessor.process(KStreamMapValues.java:40)
at org.apache.kafka.streams.processor.internals.ProcessorNode.process(ProcessorNode.java:115)
at org.apache.kafka.streams.processor.internals.ProcessorContextImpl.forward(ProcessorContextImpl.java:146)
at org.apache.kafka.streams.processor.internals.ProcessorContextImpl.forward(ProcessorContextImpl.java:129)
at org.apache.kafka.streams.processor.internals.ProcessorContextImpl.forward(ProcessorContextImpl.java:93)
at org.apache.kafka.streams.processor.internals.SourceNode.process(SourceNode.java:84)
at org.apache.kafka.streams.processor.internals.StreamTask.process(StreamTask.java:351)
at org.apache.kafka.streams.processor.internals.AssignedStreamsTasks.process(AssignedStreamsTasks.java:104)
at org.apache.kafka.streams.processor.internals.TaskManager.process(TaskManager.java:413)
at org.apache.kafka.streams.processor.internals.StreamThread.runOnce(StreamThread.java:862)
at org.apache.kafka.streams.processor.internals.StreamThread.runLoop(StreamThread.java:777)
at org.apache.kafka.streams.processor.internals.StreamThread.run(StreamThread.java:747)

I got suggestion from Thammegowda that it could be a version mismatch. So, I wanted to check what is the version of tensorflow-serving you have used so that I try to use same version.

If you feel that it isn't version problem, please let me know if I can get any insights from you to solve this issue.

Thanks in advance for your kind support.

BTW - I am able to run this on my Mac OS by creating docker using Thammegowda's github. That isn't getting supported on PPC64 system as libraries are x86 compiled.