Cloud Native Security Tools: Wiz, Lacework, Orca, and Prisma Cloud Compared
An in-depth comparison of leading CNAPP platforms — Wiz, Lacework, Orca, and Prisma Cloud — covering CSPM vs CWPP vs CNAPP, agentless vs agent-based architectures, attack path analysis, and pricing models.
The cloud security tooling market has converged around a new category: Cloud-Native Application Protection Platform (CNAPP). CNAPP combines what were previously separate product categories — CSPM, CWPP, CIEM, and CI/CD security — into a single platform. Choosing between the leading vendors requires understanding the architectural tradeoffs they've made and matching them to your security program's maturity and requirements.
Understanding the Categories
Before comparing tools, it's worth clarifying what each acronym actually means:
CSPM: Cloud Security Posture Management
CSPM tools connect to cloud provider APIs and evaluate resource configurations against security best practices and compliance frameworks. They answer: "Is my cloud configured securely?"
Key CSPM capabilities:
- Configuration assessment against CIS benchmarks, NIST, PCI-DSS, SOC2
- Drift detection (alerting when secure configurations change)
- Compliance reporting with evidence collection
- Asset inventory across cloud accounts
CSPM alone is insufficient because it only looks at configuration — it doesn't assess what's running inside VMs or containers.
CWPP: Cloud Workload Protection Platform
CWPP tools protect running workloads — VMs, containers, and serverless functions. They answer: "Are my running workloads secure and are there active threats?"
Key CWPP capabilities:
- Runtime threat detection (detecting exploitation attempts, cryptomining, lateral movement)
- Vulnerability scanning of running workloads
- Malware detection
- Process monitoring and behavioral analysis
Traditional CWPP required agents deployed in every workload, which created operational overhead.
CIEM: Cloud Infrastructure Entitlement Management
CIEM tools analyze IAM permissions and identify excessive access. They answer: "Who can access what, and is that appropriate?"
Key CIEM capabilities:
- Effective permissions analysis (what can each principal actually do?)
- Unused permission identification
- Privilege escalation path detection
- Just-in-time access recommendations
CNAPP: The Convergence
CNAPP combines CSPM + CWPP + CIEM plus:
- CI/CD pipeline security (scanning IaC, containers before deployment)
- Attack path analysis (connecting misconfigurations into exploitable paths)
- Data security posture management (finding sensitive data in cloud storage)
- Kubernetes security posture management (KSPM)
The rationale for convergence: a misconfigured security group (CSPM finding) combined with a CVE in a running container (CWPP finding) and an overpermissioned IAM role (CIEM finding) creates an attack path. Separate tools surface three separate findings; a CNAPP surfaces one prioritized risk.
Wiz: The Attack Graph Leader
Architecture
Wiz is 100% agentless. It connects via read-only API integrations to your cloud providers (AWS, Azure, GCP, OCI, Alibaba Cloud, vSphere). For workload-level visibility, Wiz uses a technique called "out-of-band scanning" — it takes snapshots of EBS volumes, disk images, and container registries and analyzes them in Wiz's own AWS environment.
The result is a graph database — the Wiz Security Graph — that models every resource, its configuration, network exposure, IAM relationships, running packages (from disk scanning), and data contents.
Attack Path Analysis
Wiz's differentiating capability is connecting individual findings into complete attack paths:
Example Wiz Attack Path:
External Internet
→ Public-facing application server (EC2 t3.large)
Exposed because: Security group allows 0.0.0.0/0:443
→ Critical CVE: CVE-2024-XXXX in application container
Severity: Critical, CVSS 9.8, exploit publicly available
→ Container has volume mount to host filesystem
Can access: host /etc, /var/lib/docker
→ Node IAM role: EC2FullAccess, S3FullAccess
Can reach: 847 S3 objects across 12 buckets
→ Sensitive data in S3 bucket "customer-exports"
Detected: 45,000 email addresses, 12,000 SSNs
Without Wiz, this would be surfaced as four separate findings:
- Security group open to internet
- Critical CVE in container
- Overpermissioned node IAM role
- PII in S3 bucket
Only by connecting them does the true risk (internet-accessible path to 45k SSNs) become visible.
DAST Integration
Wiz recently added a "Technology Preview" capability to probe exposed services and verify whether vulnerabilities are actually exploitable, reducing false positives in attack path analysis.
Pricing
Wiz pricing is workload-based:
- $15-20/month per VM
- $5-8/month per container node
- $1-3/month per serverless function
- Additional for data security features
Enterprise contracts typically include all modules. Annual contracts offer significant discounts over monthly.
Strengths and Weaknesses
Strengths:
- Fastest time-to-value (hours, not days)
- Best attack path visualization in the market
- Strong data security posture management
- No operational overhead (no agents to manage)
Weaknesses:
- No real-time runtime protection (agentless means detection delay)
- Agent required for process-level monitoring if needed
- Kubernetes visibility is good but not as deep as native Falco
Lacework: Behavioral Analytics Leader
Architecture
Lacework uses a combination of agentless cloud scanning (similar to CSPM) and an optional agent (the Lacework Agent) for deep workload monitoring. The core differentiator is the Polygraph behavior modeling system.
Polygraph works by:
- Collecting all API call activity from CloudTrail, GCP Audit Logs, Azure Monitor
- Collecting all network connections and process activity from Lacework Agents
- Building a machine learning baseline of normal behavior per account, per service, per workload
- Alerting when observed behavior deviates from the baseline
Behavioral Detection Example
Lacework's Polygraph would detect:
Anomaly: IAM Role "production-app-role" making unusual API calls
Normal behavior (last 90 days):
s3:GetObject on "app-data" bucket: 2,400/day
secretsmanager:GetSecretValue on "prod/db-password": 24/day
logs:PutLogEvents: 48,000/day
Detected anomaly (today):
ec2:RunInstances in us-west-2: 47 calls [NEW REGION - NEVER SEEN]
s3:PutObject on "attacker-bucket": 3 calls [NEW BUCKET - NEVER SEEN]
iam:CreateAccessKey: 2 calls [NEW BEHAVIOR]
Risk: HIGH - Credential compromise indicated
This detection doesn't require any rules about "RunInstances in unusual regions" — it fires purely because the behavior is different from the established baseline.
Pricing
Lacework pricing is based on compute resources monitored:
- Per vCPU/month for VM workloads (agent-based)
- Per cloud account/month for agentless scanning
- Enterprise contracts are volume-discounted
Typically $15-30/VM/month for full platform coverage.
Strengths and Weaknesses
Strengths:
- Best behavioral anomaly detection — catches novel attacks
- Low false positive rate after baseline period
- Strong threat investigation workflow
- Good coverage for financial services compliance requirements
Weaknesses:
- Baseline period (2-4 weeks) before meaningful alerts
- Less polished UI than Wiz
- Attack path visualization is less comprehensive
Orca Security: Agentless Deep Scanning
Architecture
Orca pioneered "SideScanning" — taking read-only snapshots of cloud storage (EBS, Azure Managed Disks, GCP persistent disks) and analyzing them in Orca's environment. This provides depth comparable to agent-based scanning (running processes, installed packages, sensitive files) without deploying any software.
Contextual Risk Prioritization
Orca's scoring model combines:
- Vulnerability severity (CVE CVSS score)
- Exploitability (is an exploit available? is it being actively exploited?)
- Asset exposure (is the workload internet-facing?)
- Crown jewel proximity (is sensitive data or a privileged role reachable from this workload?)
- Lateral movement potential (what other resources can be reached?)
This produces an "Orca Score" that ranks findings by actual risk rather than just technical severity. A critical CVE on an isolated dev instance scores lower than a medium CVE on an internet-facing instance with database access.
Data Security Posture Management
Orca has particularly strong data security capabilities — it scans disk snapshots for:
- PII (SSNs, credit card numbers, email addresses, phone numbers)
- Credentials (API keys, passwords, private keys)
- Regulated data patterns (HIPAA PHI, PCI cardholder data)
This catches sensitive data that leaked into unexpected locations — developer test environments with production data copies, backup buckets with unencrypted sensitive data, etc.
Pricing
Orca uses workload-based pricing similar to Wiz, typically $10-18/VM/month.
Strengths and Weaknesses
Strengths:
- True zero-agent architecture (no network access to customer workloads)
- Strong data security posture management
- Good cloud drift detection
- Fast deployment
Weaknesses:
- No real-time runtime protection
- Lower detection fidelity than agent-based tools for active threats
- Attack path analysis less mature than Wiz
Prisma Cloud (Palo Alto Networks): Most Comprehensive
Architecture
Prisma Cloud is the most feature-complete platform, built from acquisitions of RedLock (CSPM), Twistlock (container/Kubernetes security), and Bridgecrew (IaC scanning). It offers both agentless scanning and the Defender agent for deep workload protection.
Defender Agent Capabilities
The Prisma Cloud Defender agent deploys as a DaemonSet in Kubernetes and as a service on VMs. It provides:
- Real-time container runtime protection: blocking exploit attempts, alerting on suspicious process behavior
- Network firewall enforcement: enforcing microsegmentation policies at the container level
- Forensics: capturing filesystem and process state at the time of an alert for post-incident analysis
- WAAS (Web Application and API Security): in-line request inspection for web application protection
# Deploy Defenders via Helm
helm install twistlock-defender twistlock/twistlock-defender \
--namespace twistlock \
--create-namespace \
--set defender.wsAddress=wss://app.prismacloud.io:443 \
--set defender.communicationPort=8084 \
--set defender.installBundle="<token>"
CI/CD Pipeline Integration
Prisma Cloud's Bridgecrew integration scans Infrastructure as Code in CI/CD pipelines:
# GitHub Actions integration
- name: Prisma Cloud IaC Scan
uses: bridgecrewio/checkov-action@master
with:
directory: infrastructure/
check: CKV_AWS_*
framework: terraform
output_format: sarif
output_file_path: results.sarif
soft_fail: false
prisma_api_url: https://api.prismacloud.io
Pricing
Prisma Cloud pricing is complex and module-based:
- Cloud Security (CSPM): $3-8/resource/month
- Runtime Security (CWPP): $10-25/VM/month, $5-10/container/month
- Enterprise platform licensing often ~$500K-$2M/year for large organizations
Strengths and Weaknesses
Strengths:
- Most comprehensive feature set
- Best CI/CD pipeline security
- Real-time runtime protection with blocking capability
- 80+ compliance frameworks
Weaknesses:
- Most expensive option
- Longest deployment time
- Higher operational overhead
- UI can be overwhelming
Selecting the Right Tool
| Requirement | Best Fit |
|---|---|
| Fast deployment, minimal ops overhead | Wiz or Orca |
| Novel threat detection, behavioral analytics | Lacework |
| Real-time workload protection with blocking | Prisma Cloud |
| Deep data security posture management | Wiz or Orca |
| CI/CD pipeline security | Prisma Cloud (Checkov/Bridgecrew) |
| Startup / SMB budget | Orca or Lacework |
| Enterprise, regulated industry | Prisma Cloud or Wiz Enterprise |
Most security teams benefit more from choosing one platform and deploying it comprehensively than from deploying multiple tools with overlapping coverage. The exception is using a specialized tool for a specific gap — for example, using Falco for Kubernetes runtime alongside Wiz for CSPM, where Wiz doesn't provide the process-level depth needed.
The $0 option that covers significant ground: open-source tools including Trivy (vulnerability scanning), Checkov (IaC scanning), Prowler (AWS CIS benchmark), and kube-bench (Kubernetes CIS benchmark) can provide meaningful coverage before a CNAPP budget is available.