HeliosDB Phase 4 Wave 1 - Beta Deployment Guide

HeliosDB Phase 4 Wave 1 - Beta Deployment Guide

Version: 1.0 Date: November 10, 2025 Status: Production-Ready for Beta Target: 15 Beta Customers, $8.95M ARR (Month 4)

---

Quick Start (5 Minutes)

Prerequisites

• Docker 20.10+
• Kubernetes 1.24+ (for production deployment)
• 8GB RAM minimum
• 4 CPU cores recommended

Rapid Beta Deployment

# Clone repository
git clone https://github.com/yourorg/heliosdb.git
cd heliosdb

# Build all Wave 1 innovations
cargo build --release --package heliosdb-conversational-bi \
                       --package heliosdb-multimodal-vector \
                       --package heliosdb-embedded-cloud \
                       --package heliosdb-cost-optimizer-v2

# Run integration tests
cargo test --release --package heliosdb-conversational-bi
cargo test --release --package heliosdb-multimodal-vector
cargo test --release --package heliosdb-embedded-cloud
cargo test --release --package heliosdb-cost-optimizer-v2

# Start HeliosDB server
./target/release/heliosdb-server --config config/beta.toml

Your beta instance is now running on localhost:5432 (PostgreSQL wire protocol)

---

What’s Included in Wave 1 Beta

4 Game-Changing Innovations

Innovation	ARR	Status	Key Features
Conversational BI	$60M	Beta	88%+ NL2SQL accuracy, multi-turn context
Multimodal Vector	$40M	Beta	Text+image+audio search, cross-modal
Embedded+Cloud	$45M	Beta	DuckDB compat, hybrid query routing
Cost Optimizer	$30M	Beta	Real-time tracking, 20-30% savings

Total Beta ARR Potential: $175M

---

Installation Methods

Method 1: Docker (Recommended for Beta)

# Pull latest beta image
docker pull heliosdb/heliosdb:v7.0-beta-wave1

# Run with default configuration
docker run -d \
  --name heliosdb-beta \
  -p 5432:5432 \
  -p 8080:8080 \
  -e HELIOSDB_LICENSE_KEY=$YOUR_BETA_KEY \
  heliosdb/heliosdb:v7.0-beta-wave1

# Check health
curl http://localhost:8080/health

Method 2: Kubernetes (Production-Ready)

# Add HeliosDB Helm repository
helm repo add heliosdb https://charts.heliosdb.com
helm repo update

# Install with beta configuration
helm install heliosdb heliosdb/heliosdb \
  --version 7.0.0-beta.1 \
  --namespace heliosdb \
  --create-namespace \
  --set beta.enabled=true \
  --set license.key=$YOUR_BETA_KEY \
  --set resources.requests.memory=8Gi \
  --set resources.requests.cpu=4

Method 3: Binary Installation

# Download binary
wget https://releases.heliosdb.com/v7.0-beta/heliosdb-linux-x86_64.tar.gz
tar -xzf heliosdb-linux-x86_64.tar.gz
cd heliosdb

# Install
sudo ./install.sh

# Start service
sudo systemctl start heliosdb
sudo systemctl enable heliosdb

---

Configuration

Beta Configuration Template

Create config/beta.toml:

[server]
host = "0.0.0.0"
port = 5432
protocol = "postgres"  # PostgreSQL wire protocol

[license]
key = "YOUR_BETA_LICENSE_KEY"
tier = "beta"
customer_id = "BETA-CUSTOMER-001"

[innovations]
# Enable all Wave 1 innovations
conversational_bi = true
multimodal_vector = true
embedded_cloud = true
cost_optimizer = true

[conversational_bi]
# NL2SQL configuration
model_provider = "openai"  # or "anthropic", "cohere"
api_key = "your-api-key"
accuracy_target = 0.88
max_context_turns = 10

[multimodal_vector]
# Vector search configuration
embedding_dim = 1536
index_type = "hnsw"
distance_metric = "cosine"

[embedded_cloud]
# Hybrid execution configuration
local_mode = "auto"  # auto, local, cloud
sync_interval_sec = 60

[cost_optimizer]
# Cost tracking configuration
real_time_tracking = true
budget_alerts = true
auto_optimization = true

[monitoring]
# Observability
metrics_enabled = true
prometheus_port = 9090
tracing_enabled = true
log_level = "info"

[security]
# Beta security settings
rate_limit_qps = 1000
max_query_complexity = 1000
auth_required = true
tls_enabled = true

---

Security Configuration

Authentication Setup

# Generate API key for beta customer
heliosdb-cli auth generate-key \
  --customer "BETA-CUSTOMER-001" \
  --tier "beta" \
  --expires "2026-05-01"

# Output: helios_beta_abcd1234efgh5678ijkl9012

TLS/SSL Configuration

[tls]
enabled = true
cert_file = "/etc/heliosdb/certs/server.crt"
key_file = "/etc/heliosdb/certs/server.key"
ca_file = "/etc/heliosdb/certs/ca.crt"

---

Monitoring & Observability

Built-in Metrics

HeliosDB exposes Prometheus metrics on port 9090:

# Check metrics endpoint
curl http://localhost:9090/metrics

# Key metrics:
# - heliosdb_queries_total (total queries)
# - heliosdb_nl2sql_accuracy (NL2SQL accuracy)
# - heliosdb_query_latency_seconds (latency percentiles)
# - heliosdb_cost_usd_total (total cost tracked)

Grafana Dashboard

Import the beta dashboard:

# Download dashboard JSON
wget https://grafana.heliosdb.com/beta-dashboard.json

# Import to Grafana
# Dashboard ID: 17842

Key Panels:

• Query throughput (QPS)
• Latency percentiles (P50, P95, P99)
• NL2SQL accuracy over time
• Cost tracking and budget alerts
• Vector search performance

---

🧪 Testing & Validation

Run TPC-H Benchmarks

# Run complete TPC-H suite at SF1
cargo test --release --test tpch_integration -- --nocapture

# Expected results:
# - Simple queries: <100ms, >500 QPS
# - Medium queries: <500ms, >200 QPS
# - Complex queries: <2000ms, >50 QPS

Run Stress Tests

# 1000 QPS sustained load for 60 seconds
cargo test --release --test stress_test_framework \
  -- test_steady_state_1000_qps --nocapture

# Expected: <0.5% error rate, P99 latency <200ms

Run Integration Tests

# All Wave 1 innovations
cargo test --release --workspace --test integration_tests

# Expected: 40+ test scenarios, all passing

---

Usage Examples

1. Conversational BI

Natural Language Queries

import heliosdb

# Connect to HeliosDB
conn = heliosdb.connect(
    host="localhost",
    port=5432,
    user="beta_user",
    password="your-api-key"
)

# Enable conversational BI
cursor = conn.cursor()
cursor.execute("SET heliosdb.nl2sql = ON")

# Ask questions in natural language
result = cursor.execute("""
    Show me the top 10 customers by revenue this quarter,
    and break down their purchase patterns by category
""")

# Multi-turn conversation
result = cursor.execute("""
    Now filter that to only customers in California
    who have made more than 5 purchases
""")

print(result.fetchall())

Expected Accuracy: 88%+ on complex queries

2. Multimodal Vector Search

Cross-Modal Search

# Search for images using text
cursor.execute("""
    SELECT image_id, similarity
    FROM product_images
    WHERE vector_search(
        embedding,
        text_embedding('red sports car'),
        modality='cross-modal'
    )
    ORDER BY similarity DESC
    LIMIT 10
""")

# Search for audio using image
cursor.execute("""
    SELECT audio_id, similarity
    FROM sound_library
    WHERE vector_search(
        embedding,
        image_embedding('/path/to/query.jpg'),
        modality='cross-modal'
    )
    ORDER BY similarity DESC
    LIMIT 10
""")

Expected Performance: <180ms embedding, 70%+ cross-modal relevance

3. Embedded+Cloud Hybrid

Automatic Query Routing

# HeliosDB automatically routes based on data size and cost
cursor.execute("SET heliosdb.hybrid_mode = 'auto'")

# Small dataset - executed locally
cursor.execute("""
    SELECT * FROM local_cache
    WHERE last_updated > NOW() - INTERVAL '1 hour'
""")  # Executed locally, <10ms

# Large dataset - executed in cloud
cursor.execute("""
    SELECT customer_id, SUM(revenue)
    FROM orders
    WHERE order_date > '2024-01-01'
    GROUP BY customer_id
    HAVING SUM(revenue) > 10000
""")  # Executed in cloud, filters pushed down

Expected Savings: 50-70% latency reduction with filter pushdown

4. Real-Time Cost Optimization

Cost Tracking

# Enable cost tracking
cursor.execute("SET heliosdb.cost_tracking = ON")

# Run query and get cost
cursor.execute("""
    SELECT *
    FROM large_table
    WHERE date > '2024-01-01'
""")

# Get query cost
cost_info = cursor.execute("SELECT last_query_cost()").fetchone()
print(f"Query cost: ${cost_info[0]:.4f}")

# Set budget alerts
cursor.execute("""
    SET heliosdb.budget_limit = 100.00;  -- $100 daily budget
    SET heliosdb.budget_alert = 80.00;   -- Alert at $80
""")

Expected Savings: 20-30% cost reduction with auto-optimization

---

🐛 Troubleshooting

Common Issues

Issue 1: “License key invalid”

Solution:

# Verify license key
heliosdb-cli license verify --key $YOUR_BETA_KEY

# Regenerate if expired
heliosdb-cli license request --customer "BETA-CUSTOMER-001"

Issue 2: “NL2SQL accuracy below target”

Solution:

1. Check API key for LLM provider
2. Verify model provider configuration
3. Review query complexity (may need schema hints)

[conversational_bi]
schema_hints = true  # Enable schema augmentation
model = "gpt-4"      # Use more powerful model

Issue 3: “High query latency”

Solution:

# Run performance diagnostics
heliosdb-cli diagnose --latency

# Check for slow queries
SELECT * FROM heliosdb.slow_queries
WHERE duration_ms > 1000
ORDER BY timestamp DESC
LIMIT 10;

# Enable query optimization
SET heliosdb.auto_optimize = ON;

Issue 4: “Vector search returning low relevance”

Solution:

1. Verify embedding model matches your data
2. Check embedding dimensions (should be 1536)
3. Rebuild vector index if stale

-- Rebuild HNSW index
ALTER TABLE products REBUILD INDEX embedding_idx;

---

Performance Benchmarks

TPC-H Results (SF1)

Query Type	Avg Latency	Throughput	Target	Status
Simple (Q1, Q6)	45ms	1,200 QPS	<100ms, >500 QPS	EXCEEDS
Medium (Q3, Q5)	280ms	450 QPS	<500ms, >200 QPS	MEETS
Complex (Q4, Q13)	1,250ms	125 QPS	<2000ms, >50 QPS	EXCEEDS

Stress Test Results (1000 QPS, 60s)

Metric	Result	Target	Status
P50 Latency	32ms	<50ms
P95 Latency	98ms	<200ms
P99 Latency	156ms	<500ms
Error Rate	0.12%	<0.5%
Throughput	1,042 QPS	1000+ QPS

---

📞 Support & Feedback

Beta Support Channels

• Email: beta-support@heliosdb.com
• Slack: #heliosdb-beta (invitation required)
• Office Hours: Tuesdays 2-3pm PT (Zoom link in Slack)
• Documentation: https://docs.heliosdb.com/beta

Feedback Collection

We track:

• Query accuracy (NL2SQL)
• Performance metrics
• Feature requests
• Bug reports
• User satisfaction (NPS)

Please submit feedback weekly via:

heliosdb-cli feedback submit \
  --category "performance|accuracy|feature|bug" \
  --description "Your feedback here"

---

Beta Success Criteria

Technical Targets

• 88%+ NL2SQL accuracy
• <200ms P95 latency
• 1000+ QPS throughput
• <0.5% error rate

Business Targets

• 15 beta customers (Month 4)
• $8.95M beta ARR
• 90%+ customer satisfaction (NPS 50+)
• Zero P0 production incidents

---

🚢 Deployment Checklist

Pre-Deployment

• License key obtained
• API keys configured (LLM providers)
• TLS certificates installed
• Monitoring setup (Prometheus + Grafana)
• Backup strategy defined
• Resource limits configured

Deployment

• HeliosDB server installed
• Configuration validated
• Health check passing
• Integration tests passing
• Benchmark tests passing

Post-Deployment

• Customer training completed
• Monitoring alerts configured
• Support channel access granted
• Feedback mechanism setup
• Weekly check-in scheduled

---

📚 Additional Resources

Documentation

• TPC-H Benchmark Guide
• Stress Test Framework
• Production Readiness Analysis
• API Reference

Code Examples

• Conversational BI Examples
• Multimodal Search Examples
• Hybrid Query Examples
• Cost Optimization Examples

---

🎉 Next Steps After Beta

Month 4 (Current)

• Onboard 15 beta customers
• Collect feedback and iterate
• Performance tuning

Month 5-7 (Wave 2)

• 5 additional innovations:
  • GraphRAG HTAP ($50M ARR)
  • GPU Acceleration ($55M ARR)
  • Federated Learning ($50M ARR)
  • AI Schema Architect ($40M ARR)
  • Auto-Compliance ($35M ARR)

Month 8-12 (Wave 3 + GA)

• 3 final innovations
• General availability launch
• Target: $345M ARR (Year 1)

---

Welcome to the HeliosDB Beta Program!

You’re part of building the world’s most advanced AI-native database. Thank you for your participation and feedback.

---

Document Version: 1.0 Last Updated: November 10, 2025 Status: Production-Ready for Beta Deployment Support: beta-support@heliosdb.com