ika

Run

Run via Profile

Profile is configured in conf/application.conf (or conf/application-${SITE}.conf):

pool_1 {
  type="pool://"
  strategy="lb"
  destinations = [
    "host1=http://localhost:8300",
    "host2=http://localhost:8301"
  ]
}

retry_2 {
  type="retry://"
  maxRetries=3
  delay=1000
}

http_2 {
  type="http://"
  connectTimeout=1000
  responseTimeout=3000
}

profiles {    
  
  proxy3 = {
    processors = "pool_1, retry_2, http_2"
  }
  
}

Running with profile:

./run-ika.sh --profile proxy

Run via pipeline

Run with cache processor and expiration 12 seconds to the Pool processor with 1 node

./run-ika.sh cache://120000 pool://1=http://geth:8545

Run with 4 Threads pool, gzip compression processor to the pool processor with 2 nodes

./run-ika.sh --threads=4 compress://gzip pool://http://geth-1:8545,http://geth-2:8545

Run with Round-robin load-balancing processor

./run-ika.sh lb://tag1=http://geth-1:8545,tag2=http://geth-2:8545

Run with RPC3 aware processor (rpc3:// special cache which knows how to parse batches and use cache and exceptions)

./run-ika.sh rpc3://evm,10000,block  pool://tag1=http://geth-1:8545,tag2=http://geth-2:8545

Run with AI processors: ai:// processor knows how to route to the model, ai_tokens:// knows how to calculate tokens from response

./run-ika.sh rpc3://evm,10000,block  pool://tag1=http://geth-1:8545,tag2=http://geth-2:8545