Saltar al contenido
Fali Fuentes

AWS makes Bedrock AgentCore runtime V2 default and publishes self-healing ETL guide

Leer este artículo en español

AWS has changed the default runtime of Amazon Bedrock AgentCore to version V2 in 22 regions —including Europe (Spain, Frankfurt, Ireland, London, Milan, Paris), the U.S., Canada, South America (São Paulo), and Asia Pacific— starting November 12, 2026. Existing runtimes remain on V1 unless you explicitly update them with platformVersion=V2; if you need to force V1 in a region where V2 is the default, use platformVersion=V1 when creating or updating the runtime.

Runtime V2 default in 22 regions since November 12

The official announcement confirms that the new generation of AgentCore runtime is now the default in the listed regions. Each session is restored from a snapshot of the initialized agent, eliminating the traditional cold start and allowing compute and memory to scale with the workload. Inactive agents scale to zero, and memory is reclaimed within minutes.

Startup in <2 s, scale-to-zero, and up to 68% less billed memory

With 2 GB images, the P75 cold start drops to under 2 seconds compared to 29.7 s in V1. Scaling reaches up to 8 vCPU, 32 GB of memory, and images up to 10 GB. AWS quantifies up to 68% less billed memory for multi-turn agents with natural pauses; for continuous streaming agents the savings will be lower. Monitor actual cost with Cost Explorer filtering by AgentCoreRuntime.

Controlled migration: you decide when to move each runtime

Existing runtimes do not migrate automatically. To move one, update it with platformVersion=V2. In regions where V2 is not yet the default (Middle East, Israel, China), you must also specify platformVersion=V2 when creating it. Validate in staging with your real images before touching production: measure cold start, memory reclamation, and billing.

Self-healing ETL pattern: detect→correct→persist→remediate loop

The next day, the AWS Big Data blog published a complete guide that implements a closed-loop data quality. The architecture uses AWS Glue as a quality gate and Bedrock AgentCore (Strands Agents SDK) to orchestrate four agents: orchestrator, correction, persistence and remediation. Correction combines semantic search in Knowledge Bases with deterministic lookup in S3 Vectors.

Four agents, dual correction, and configurable confidence threshold

The orchestrator only applies the correction if the strategy exceeds the 0.7 confidence threshold; otherwise it notifies via SNS for manual review. You can adjust the threshold by domain: 0.85+ for financial or healthcare data, 0.6 for non-sensitive telemetry, always with the SNS fallback active. The optional remediation opens pull requests in GitHub via MCP Server with the root cause, but does not auto-merge.

End-to-end CDK deployment and integrated governance

The example repository deploys six CDK Python stacks (DataLake, Ingestion, Glue, KnowledgeBase, AgentCore, AgentCoreRuntime) and includes test scripts with synthetic NOAA data. The zone corrected/ of the data lake is audited with metadata from the applied strategy. You can enable Bedrock Guardrails as an extra governance layer.

Immediate steps for your environment

  • Validate V2 in staging with your real images: measure cold-start, memory reclamation, and real cost with Cost Explorer filtering by AgentCoreRuntime.
  • If you are in an unlisted region (Middle East, Israel, China), deploy with platformVersion=V2 explicitly and verify availability before architecting multi-region.
  • For automatic remediation, use Secrets Manager and GitHub MCP Server with a token of scope repo and rotated credentials; the agent only opens PRs, does not merge.
  • Adjust the confidence threshold based on the criticality of your data and always keep the human fallback via SNS active.

Structured reference for readers and AI agents

{
  "format": "rafael-fuentes-article/1.0",
  "language": "en",
  "title": "AWS makes Bedrock AgentCore runtime V2 default and publishes self-healing ETL guide",
  "summary": "Starting November 12, new AgentCore runtimes use V2 in 22 regions with <2 s startup and scale-to-zero; AWS publishes a self-healing ETL pattern with four agents and dual correction.",
  "author": "Rafael Fuentes",
  "publicationDate": "2026-11-18T06:01:16.052+01:00",
  "eventDate": "2026-11-12",
  "url": "https://falifuentes.com/rafael-fuentes-2026-11-18-am-en/",
  "translation": {
    "language": "es",
    "url": "https://falifuentes.com/rafael-fuentes-2026-11-18-am-es/"
  },
  "topics": [
    "AI",
    "Artificial Intelligence",
    "English"
  ],
  "sources": [
    "https://aws.amazon.com/about-aws/whats-new/2026/11/new-agentcore-runtime-default",
    "https://aws.amazon.com/blogs/big-data/build-a-self-healing-etl-pipeline-with-amazon-bedrock-agentcore"
  ]
}
SYSTEM_EXPERT
Rafael Fuentes – BIO

I am a seasoned cybersecurity expert with over twenty years of experience leading strategic projects in the industry. Throughout my career, I have specialized in comprehensive cybersecurity risk management, advanced data protection, and effective incident response. I hold a certification in Industrial Cybersecurity, which has provided me with deep expertise in compliance with critical cybersecurity regulations and standards. My experience includes the implementation of robust security policies tailored to the specific needs of each organization, ensuring a secure and resilient digital environment.

Share
Scroll al inicio