minor changes

Author: alonre24

docs/performance.md

@@ -18,7 +18,7 @@ This page is intended to provide clarity on how to obtain the benchmark numbers
 ---------------------------------------
 
 ## Methodology and Infrastructure checklist
-As stated previously we encourage the community to run and the benchmarks on their own infrastructure and specific use case. As part of a reproducible and stable metodology we recommend that for each tested version/variation:
+As stated previously we encourage the community to run the benchmarks on their own infrastructure and specific use case. As part of a reproducible and stable metodology we recommend that for each tested version/variation:
 
 - Monitoring should be used to assert that the machines running the the benchmark client do not become the performance bottleneck.
 
@@ -42,10 +42,10 @@ The following example will:
 - Enable multi-threaded client and use 2 benchmark client threads, meaning the benchmark client maximum CPU usage will be 200%.
 - Be executed without pipelining (pipeline value of 1).
 - If server replies with errors, show them on stdout.
-- Benchmark [AI.MODELRUN](https://oss.redislabs.com/redisai/commands/#aimodelrun) command a model stored as a key's value using its specified backend and device.
+- Benchmark [AI.MODELEXECUTE](https://oss.redislabs.com/redisai/commands/#aimodelexecute) command over a model stored as a key's value using its specified backend and device.
 
 ```
-redis-benchmark -e --threads 2 -c 50 -P 1 -n 100000 AI.MODELRUN <key> INPUTS <input> [input ...] OUTPUTS <output> [output ...]
+redis-benchmark --threads 2 -c 50 -P 1 -n 100000 AI.MODELEXECUTE <key> INPUTS <input_count> <input> [input ...] OUTPUTS <output_count> <output> [output ...]
 ```
 
 ## Using memtier-benchmark
@@ -59,22 +59,22 @@ The following example will:
 - Store the benchmark result in `result.json`.
 - Be executed without pipelining (pipeline value of 1).
 - If server replies with errors, show them on stdout.
-- Benchmark both [AI.MODELRUN](https://oss.redislabs.com/redisai/commands/#aimodelrun) command a model stored as a key's value using its specified backend and device, and [AI.SCRIPTRUN](https://oss.redislabs.com/redisai/commands/#aiscriptrun) stored as a key's value on its specified device.
+- Benchmark both [AI.MODELEXECUTE](https://oss.redislabs.com/redisai/commands/#aimodelexecute) command a model stored as a key's value using its specified backend and device, and [AI.SCRIPTEXECUTE](https://oss.redislabs.com/redisai/commands/#aiscriptexecute) stored as a key's value on its specified device.
 
 ```
-memtier_benchmark --clients 50 --threads 4 --requests 10000 --pipeline 1 --json-out-file results.json --command "AI.MODELRUN model_key INPUTS input1 OUTPUTS output1" --command "AI.SCRIPTRUN script_key INPUTS input2 OUTPUTS output2"
+memtier_benchmark --clients 50 --threads 4 --requests 10000 --pipeline 1 --json-out-file results.json --command "AI.MODELEXECUTE model_key INPUTS input_count input1 ... OUTPUTS output_count output1 ..." --command "AI.SCRIPTEXECUTE script_key entry_point INPUTS input_count input1 ... OUTPUTS output_count output1 ..."
 ```
 
 ## Using aibench
 
 _AIBench_ is a collection of Go programs that are used to generate datasets and then benchmark the inference performance of various Model Servers. The intent is to make the AIBench extensible so that a variety of use cases and Model Servers can be included and benchmarked.
 
 
-We recommend that you follow the detailed install steps [here](https://github.com/RedisAI/aibench#installation) and refer to the per-use [documentation](https://github.com/RedisAI/aibench#current-use-cases).
+We recommend that you follow the detailed installation steps [here](https://github.com/RedisAI/aibench#installation) and refer to the per-use [documentation](https://github.com/RedisAI/aibench#current-use-cases).
 
 ###  Current DL solutions supported:
 
-- [RedisAI](https://redisai.io): an AI serving engine for real-time applications built by Redis Labs and Tensorwerk, seamlessly plugged into ​Redis.
+- [RedisAI](https://redisai.io): an AI serving engine for real-time applications built by Redis and Tensorwerk, seamlessly plugged into ​Redis.
 - [Nvidia Triton Inference Server](https://docs.nvidia.com/deeplearning/triton-inference-server): An open source inference serving software that lets teams deploy trained AI models from any framework (TensorFlow, TensorRT, PyTorch, ONNX Runtime, or a custom framework), from local storage or Google Cloud Platform or AWS S3 on any GPU- or CPU-based infrastructure.
 - [TorchServe](https://pytorch.org/serve/): built and maintained by Amazon Web Services (AWS) in collaboration with Facebook, TorchServe is available as part of the PyTorch open-source project.
 - [Tensorflow Serving](https://www.tensorflow.org/tfx/guide/serving): a high-performance serving system, wrapping TensorFlow and maintained by Google.