Adding Data Cache

Signed-off-by: Akshay Chitneni <achitneni@apple.com>

Author: Akshay Chitneni

.gitignore

@@ -26,3 +26,8 @@ cover.out
 # Helm
 charts/kubeflow-trainer/charts/
 charts/kubeflow-trainer/Chart.lock
+
+# data_cache
+pkg/data_cache/target
+pkg/data_cache/src/client/target/
+hack/data_cache/*.log

cmd/data_cache/Dockerfile

@@ -0,0 +1,50 @@
+FROM rust:1.85-bullseye AS builder
+
+WORKDIR /usr/src/arrow_cache
+
+ENV RUST_LOG=info
+
+# Install system dependencies
+RUN apt-get update && \
+    apt-get -y install libssl-dev openssl zlib1g zlib1g-dev libpq-dev cmake protobuf-compiler netcat curl && \
+    rm -rf /var/lib/apt/lists/*
+
+# Install cargo-chef for better caching
+RUN rustup update && cargo install cargo-chef --version 0.1.62
+
+# Copy manifests for dependency caching
+COPY pkg/data_cache/Cargo.toml pkg/data_cache/Cargo.lock ./
+
+# Cache dependencies
+RUN cargo fetch
+
+# Copy source code
+COPY pkg/data_cache/ .
+
+# Run tests
+RUN cargo test --tests
+
+# Build binaries in release mode
+RUN cargo build --release --bin head --bin worker
+
+# Stage 2: Create a minimal runtime image
+FROM debian:bookworm-slim AS runtime
+
+# Install runtime dependencies
+RUN apt-get update && \
+    apt-get -y install ca-certificates && \
+    rm -rf /var/lib/apt/lists/*
+
+# Copy binaries from builder stage
+COPY --from=builder /usr/src/arrow_cache/target/release/head /usr/local/bin/head
+COPY --from=builder /usr/src/arrow_cache/target/release/worker /usr/local/bin/worker
+
+# Create non-root user
+RUN groupadd -r cache_user && useradd -r -g cache_user cache_user
+
+# Change ownership and switch to non-root user
+RUN chown -R cache_user:cache_user /usr/local/bin/
+USER cache_user
+
+# Set default command
+CMD ["head"]

hack/data_cache/run_with_remote_table.sh

@@ -0,0 +1,88 @@
+#!/bin/bash
+
+# Check for required arguments
+if [ $# -lt 5 ]; then
+    echo "Usage: $0 <iam-role-arn> <metadata-loc> <table-name> <schema-name> <aws-profile> [environment]"
+    exit 1
+fi
+
+IAM_ROLE_ARN="$1"
+METADATA_LOC="$2"
+TABLE_NAME="$3"
+SCHEMA_NAME="$4"
+AWS_PROFILE="$5"
+ENVIRONMENT="${6:-LOCAL}"
+
+echo "Using IAM Role: $IAM_ROLE_ARN"
+echo "Metadata Location: $METADATA_LOC"
+echo "Table Name: $TABLE_NAME"
+echo "Schema Name: $SCHEMA_NAME"
+echo "Environment: $ENVIRONMENT"
+echo "AWS Profile: $AWS_PROFILE"
+
+role_output=$(aws sts assume-role --role-arn "$IAM_ROLE_ARN" --role-session-name "RoleSession1" --profile "$AWS_PROFILE")
+
+# Parse the JSON output using jq
+export AWS_ACCESS_KEY_ID=$(echo $role_output | jq -r '.Credentials.AccessKeyId')
+export AWS_SECRET_ACCESS_KEY=$(echo $role_output | jq -r '.Credentials.SecretAccessKey')
+export AWS_SESSION_TOKEN=$(echo $role_output | jq -r '.Credentials.SessionToken')
+export AWS_REGION=us-west-2
+export AWS_EC2_METADATA_DISABLED=true
+
+# Set required environment variables for testing
+export METADATA_LOC="$METADATA_LOC"
+export TABLE_NAME="$TABLE_NAME"
+export SCHEMA_NAME="$SCHEMA_NAME"
+export RUNTIME_ENV="$ENVIRONMENT"
+
+# Function to cleanup processes on exit
+cleanup() {
+    echo ""
+    echo "Stopping services..."
+    kill -9 $WORKER1_PID $WORKER2_PID $HEAD_PID 2>/dev/null || true
+    wait $WORKER1_PID $WORKER2_PID $HEAD_PID 2>/dev/null || true
+    rm -rf /tmp/test_metadata
+    exit 0
+}
+
+# Set up signal handlers for graceful shutdown
+trap cleanup SIGINT SIGTERM
+
+# Function to check if a service is ready
+check_service_ready() {
+    local host=$1
+    local port=$2
+    local service_name=$3
+    
+    echo "Waiting for $service_name to be ready on $host:$port..."
+    while ! nc -z "$host" "$port" 2>/dev/null; do
+        echo "  $service_name not ready yet, waiting 2 seconds..."
+        sleep 2
+    done
+    echo "  $service_name is ready!"
+}
+
+echo "Starting worker node 1..."
+cargo run --bin worker -- 0.0.0.0 50052 > worker1.log 2>&1 &
+WORKER1_PID=$!
+
+echo "Starting worker node 2..."
+cargo run --bin worker -- 0.0.0.0 50053 > worker2.log 2>&1 &
+WORKER2_PID=$!
+
+# Wait for both workers to be ready
+check_service_ready localhost 50052 "worker1"
+check_service_ready localhost 50053 "worker2"
+
+echo "Both workers are ready, starting head node..."
+cargo run --bin head -- 0.0.0.0 50051 > head.log 2>&1 &
+HEAD_PID=$!
+
+check_service_ready localhost 50051 "head"
+
+echo "All services are running. Press Ctrl+C to stop all services."
+wait
+
+#echo "Running client test..."
+#cd src/client && cargo run 2>&1
+#CLIENT_EXIT_CODE=$?